353 docs tagged with "llm"

Test-time scaling for complex reasoning tasks shows that leveraging inference-time compute, by methods such as independently sampling and aggregating mu...

Conversational agents are increasingly deployed in knowledge-intensive settings, where correct behavior depends on retrieving and applying domain-specif...

Large language models are beginning to show steganographic capabilities. Such capabilities could allow misaligned models to evade oversight mechanisms....

Emotion Recognition in Conversations (ERC) presents unique challenges, requiring models to capture the temporal flow of multi-turn dialogues and to effe...

Large language model (LLM) agents, such as OpenAI's Operator and Claude's Computer Use, can automate workflows but unable to handle payment tasks. Exist...

Research in AI using Large-Language Models (LLMs) is rapidly evolving, and the comparison of their performance with human reasoning has become a key con...

Under the lens of Marr's levels of analysis, we critique and extend two claims about language models (LMs) and language processing: first, that predicti...

Large vision--language models (VLMs) are increasingly applied to long-video question answering, yet inference is often bottlenecked by the number of inp...

We study adaptive querying for learning user-dependent quantities of interest, such as responses to held-out items and psychometric indicators, within t...

Go beyond naive RAG - master query transformation, HyDE, multi-query retrieval, Self-RAG, Corrective RAG, and iterative retrieval patterns for complex questions.

Transformer attention is typically implemented using softmax normalization, which enforces attention weights with unit sum normalization. While effectiv...

Measuring LLM agent performance through trajectory analysis, benchmark suites, LLM-as-judge, failure taxonomies, and production monitoring strategies.

Implementing defense-in-depth safety for production LLM agents - prompt injection defense, input/output guardrails, tool sandboxing, HITL confirmation, and audit logging.

While Multi-Agent Systems (MAS) excel in complex reasoning, they suffer from the cascading impact of erroneous information generated by individual parti...

Translating natural language into Jira Query Language (JQL) requires resolving ambiguous field references, instance-specific categorical values, and com...

Build agents that control their own retrieval - multi-step reasoning, router agents, ReAct loops, LangGraph stateful pipelines, and production patterns for agentic retrieval systems.

The expansion of retrieval-augmented generation (RAG) into multimodal domains has intensified the challenge for processing complex visual documents, suc...

Deep Research agents are rapidly emerging as primary consumers of modern retrieval systems. Unlike human users who issue and refine queries without docu...

Safety benchmarks, capability evaluations, LLM judges, uplift assessments, and how labs like Anthropic use evaluation-gated deployment through Responsible Scaling Policies.

We present a winning three-stage system for SemEval 2026 Task~12: Abductive Event Reasoning that combines graph-based retrieval, LLM-driven abductive re...

Explainable AI (XAI) research has experienced substantial growth in recent years. Existing XAI methods, however, have been criticized for being technica...

Direct Preference Optimization (DPO) is widely used after supervised fine-tuning (SFT) to align language models, yet empirical behavior under small back...

Kimi K2.5 is an open-weight LLM that rivals closed models across coding, multimodal, and agentic benchmarks, but was released without an accompanying sa...

Argumentative LLMs (ArgLLMs) are an existing approach leveraging Large Language Models (LLMs) and computational argumentation for decision-making, with...

Can narratives make arguments more persuasive? And to this end, which narrative features matter most? Although stories are often seen as powerful tools...

The rapid advancement of large language models (LLMs) has enabled powerful authorship inference capabilities, raising growing concerns about unintended...

Large language models (LLMs) increasingly rely on chain-of-thought (CoT) reasoning to solve complex tasks. Yet ensuring that the reasoning trace both co...

The 2017 Vaswani et al. paper that replaced recurrent networks with pure attention - and why it changed everything.

How modern AI systems process speech and audio - from Whisper's spectrogram-based ASR to end-to-end audio understanding in GPT-4o and Gemini.

This paper studies Automated Instruction Revision (AIR), a rule-induction-based method for adapting large language models (LLMs) to downstream tasks usi...

Understand how LLMs generate tokens one at a time, why decoding is memory-bandwidth bound, and how to reason about inference latency with the roofline model.

Large language models have shown strong performance on broad-domain knowledge and reasoning benchmarks, but it remains unclear how well language models...

Large language models (LLMs) often produce confident but incorrect answers in settings where abstention would be safer. Standard evaluation protocols, h...

Navigate the LLM benchmark ecosystem - what each benchmark actually measures, saturation, contamination, and how to build benchmarks that can't be gamed.

Accurate evaluation is central to the large language model (LLM) ecosystem, guiding model selection and downstream adoption across diverse use cases. In...

Large language models (LLMs) are becoming increasingly capable mathematical collaborators, but static benchmarks are no longer sufficient for evaluating...

Evaluating the factuality of long-form output generated by large language models (LLMs) remains challenging, particularly when responses are open-ended...

Recent advances in multimodal Retrieval-Augmented Generation (RAG) enable Large Language Models (LLMs) to analyze enterprise spreadsheet workbooks conta...

Large language models are increasingly deployed in settings where reliability matters, yet output-level uncertainty signals such as token probabilities,...

Large language models (LLMs) encode vast world knowledge in their parameters, yet they remain fundamentally limited by static knowledge, finite context...

Large language models with web search are increasingly used in scientific publishing agents, yet they still produce BibTeX entries with pervasive field-...

Master reference-based generation metrics - BLEU, ROUGE, BERTScore, BLEURT - and know exactly when each one lies to you.

Four caching layers for LLM applications - exact match, semantic similarity, provider prefix caching, and KV cache - with implementation patterns and production tradeoffs.

Production patterns for calling LLM APIs - authentication, retry logic, rate limiting, error handling, async calls, and the Anthropic and OpenAI Python SDKs.

Large language models are increasingly deployed as autonomous coding agents and have achieved remarkably strong performance on software engineering benc...

Firearm violence is a pressing public health issue, yet research into survivors' lived experiences remains underfunded and difficult to scale. Qualitati...

Five detailed production LLM architectures - GitHub Copilot, Notion AI, customer support bots, enterprise RAG, and code review agents - with real architecture decisions, scale numbers, and lessons learned.

Evidence-grounded reasoning requires more than attaching retrieved text to a prediction: a model should make decisions that depend on whether the provid...

Learn how GPT-style autoregressive models work, the evolution from GPT-1 to GPT-4, sampling strategies, and why causal LM became the dominant paradigm for LLMs.

Large language models (LLMs) are trained on enormous amounts of data and encode knowledge in their parameters. We propose a pipeline to elicit causal re...

Learn how to unlock multi-step reasoning in LLMs by making them think out loud - and why this simple technique dramatically improves accuracy on complex tasks.

How chain-of-thought prompting transforms model reasoning - from the Wei et al. 2022 breakthrough to self-consistency, process supervision, and the faithfulness problem.

The transformer is the most popular neural architecture for language modeling. The cornerstone of the transformer is its global attention mechanism, whi...

Large language models (LLMs) have created new opportunities to enhance the efficiency of scholarly activities; however, challenges persist in the ethica...

This paper reports on the development of a leaderboard of Open Large Language Models (LLM) for European Portuguese (PT-PT), and on its associated benchm...

How CLIP uses contrastive learning on 400M image-text pairs to build a shared semantic embedding space - enabling zero-shot classification without labeled data.

How domain-specific pre-training and fine-tuning on code and math data produces models that outperform general LLMs on programming and reasoning tasks - and when to use them in production.

The rapid adoption of Large Language Models (LLMs) has transformed modern software development by enabling automated code generation at scale. While the...

Activation steering methods enable inference-time control of large language model (LLM) behavior without retraining, but current approaches face a funda...

Mobile Agents can autonomously execute user instructions, which requires hybrid-capabilities reasoning, including screen summary, subtask planning, acti...

How Anthropic replaced human feedback with AI feedback guided by explicit principles - the Constitutional AI technique, RLAIF, and how it enables scalable alignment.

The mathematics of constrained decoding - finite-state machines, token masking, context-free grammars, and how the Outlines library achieves guaranteed JSON schema conformance at generation time.

How LLMLingua, AutoCompressors, GIST tokens, and selective compression reduce long contexts to fewer tokens while preserving the information needed to answer queries.

How position interpolation, NTK-aware scaling, YaRN, and LongLoRA extend pretrained models to context windows far beyond their original training length.

Engineering strategies for managing context windows in production LLM applications - history truncation, compression, RAG ordering, and prompt caching design.

Speech foundation models struggle with low-resource Pacific Indigenous languages because of severe data scarcity. Furthermore, full fine-tuning risks ca...

Learn how continuous batching eliminates GPU idle time by replacing finished sequences immediately rather than waiting for the longest request in a batch to complete.

AI agents powered by reasoning models require access to sensitive user data. However, their reasoning traces are difficult to control, which can result...

Large language models (LLMs) can struggle to memorize factual knowledge in their parameters, often leading to hallucinations and poor performance on kno...

We aim to examine the extent to which Large Language Models (LLMs) can 'talk much' about grammar modules, providing evidence from syntax core properties...

How DARE randomly drops delta weights and rescales the remainder to dramatically reduce interference when merging multiple fine-tuned models.

The fast-growing demands in using Large Language Models (LLMs) to tackle complex multi-step data science tasks create an emergent need for accurate benc...

Large Language Model (LLM) adapters enable low-cost model specialization, but introduce complex caching and scheduling challenges in distributed serving...

Training capable software engineering (SWE) agents demands large-scale, executable, and verifiable environments that provide dynamic feedback loops for...

DeepSeek's innovations in mixture of experts - fine-grained experts, shared experts, DeepSeek-V2 and V3, multi-token prediction, and training for $6M.

How DeepSeek built an open-weights reasoning model using pure RL with GRPO, the R1-Zero experiment, distillation into smaller models, and what open-source reasoning means for the research community.

Large language models and deep research agents supply citation URLs to support their claims, yet the reliability of these citations has not been systema...

Reinforcement learning with verifiable rewards (RLVR) typically optimizes for outcome rewards without imposing constraints on intermediate reasoning. Th...

The ability to provide trustworthy maternal health information using phone-based chatbots can have a significant impact, particularly in low-resource se...

How denoising diffusion models learn to reverse a Gaussian noise process to generate high-quality images from text prompts.

The language in online platforms, influence operations, and political rhetoric frequently directs a mix of pro-social sentiment (e.g., advocacy, helpful...

Achieving human-like responsiveness is a critical yet challenging goal for cascaded spoken dialogue systems. Conventional ASR-LLM-TTS pipelines follow a...

Multi-turn interactions with large language models typically retain the assistant's own past responses in the conversation history. In this work, we rev...

Reasoning in vision-language models (VLMs) has recently attracted significant attention due to its broad applicability across diverse downstream tasks....

Master the art and science of splitting documents into chunks that maximize retrieval precision - the most underestimated decision in RAG system design.

Automated annotation of pedagogical dialogue is a high-stakes task where LLMs often fail without sufficient domain grounding. We present a domain-adapte...

Direct Preference Optimization and its successors - how DPO eliminates the need for a separate reward model and RL training, plus IPO, KTO, SimPO, and ORPO.

Master DPO - the elegant insight that you can optimize LLMs for human preferences without training a reward model or running RL, derived directly from the optimal RLHF policy.

Adapting Large Language Models (LLMs) to specialized domains requires high-quality instruction tuning datasets, which are expensive to create through hu...

Multimodal web agents that process both screenshots and accessibility trees are increasingly deployed to interact with web interfaces, yet their dual-st...

The syntactic structure of a sentence can be represented as a tree where edges indicate syntactic dependencies between words. When that structure is a s...

Text-to-SQL enables non-expert users to query databases in natural language, yet real-world schemas often suffer from ambiguous, abbreviated, or inconsi...

A comprehensive survey of the embedding model ecosystem - SBERT, contrastive learning, SimCSE, E5, BGE, GTE, OpenAI, Voyage AI, Cohere, and the MTEB leaderboard.

Master embedding model selection for retrieval - MTEB benchmarks, model families, Matryoshka embeddings, bi-encoders vs cross-encoders, and fine-tuning strategies.

Reducing embedding storage and search costs - float32 to float16, int8, and binary quantization, Hamming distance search, the rescoring trick, and implementation with FAISS and Qdrant.

How token embeddings form a dense vector space that captures semantic meaning - geometry, anisotropy, weight tying, and visualization.

Build, deploy, and operate production-grade embedding pipelines - caching, incremental indexing, staleness management, vector DB selection, and cost optimization at scale.

Comparing encoder-only, decoder-only, and encoder-decoder transformer architectures - when to use each and why decoder-only won.

The Hyperspace Analogue to Language (HAL) model relies on global word co-occurrence matrices to construct distributional semantic representations. While...

As corporate responsibility increasingly incorporates environmental, social, and governance (ESG) criteria, ESG reporting is becoming a legal requiremen...

The EU AI Act, US executive orders, UK AI policy, China AI regulations, and practical compliance implications for AI engineers building and deploying language models.

MTEB benchmark deep dive, nDCG@10, Recall@K, MRR, MAP, building domain-specific evaluation sets, running MTEB locally, and avoiding the contamination problem.

In contested domains, instruction-tuned language models must balance user-alignment pressures against faithfulness to the in-context evidence. To evalua...

The benchmark landscape for reasoning models - AIME, MATH-500, Codeforces, ARC-AGI, GPQA Diamond, process vs. outcome evaluation, and contamination concerns.

As large language models (LLMs) advance in linguistic competence, their reasoning abilities are gaining increasing attention. In humans, reasoning often...

Reinforcement learning (RL) has become essential to the post-training of large language models (LLMs) for reasoning, agentic capabilities and alignment....

The role of position-wise feed-forward networks in transformers - from the basic FFN to SwiGLU and Mixture of Experts.

Master in-context learning by providing carefully selected examples that demonstrate the exact behavior you want - without any model fine-tuning.

Contrastive fine-tuning with triplet loss, hard negative mining, in-batch negatives, synthetic data generation, TSDAE, GPL, and a full worked example on domain adaptation.

Transformer-based large language models exhibit in-context learning, enabling adaptation to downstream tasks via few-shot prompting with demonstrations....

Large language models (LLMs) are increasingly applied in financial scenarios. However, they may produce harmful outputs, including facilitating illegal...

Layer grafting, depth upscaling, Solar 10.7B, and the fundamental limits of what model merging can and cannot achieve.

We present frequency-ordered tokenization, a simple preprocessing technique that improves lossless text compression by exploiting the power-law frequenc...

The complexity of Vietnam's legal texts presents a significant barrier to public access to justice. While Large Language Models offer a promising soluti...

Large language models (LLMs) have recently reshaped Automated Essay Scoring (AES), yet prior studies typically examine individual techniques in isolatio...

Reinforcement learning (RL) for large language models (LLMs) increasingly relies on sparse, outcome-level rewards -- yet determining which actions withi...

A practical decision framework for choosing between full fine-tuning, LoRA, QLoRA, prompt tuning, and other PEFT methods based on your model size, data, and quality requirements.

Master Microsoft's Graph RAG - build knowledge graphs from documents, use community detection for global queries, and understand when graph structure beats flat vector search.

Build layered defense-in-depth safety systems for LLM applications - input filtering, toxicity detection, PII redaction, prompt injection defense, output validation, and human review escalation.

We present H-RAG, our submission to SemEval-2026 Task 8 (MTRAGEval), addressing both Task A (Retrieval) and Task C (Generation with Retrieved Passages)....

Cyberbullying on social media is inherently multilingual and multi-faceted, where abusive behaviors often overlap across multiple categories. Existing m...

Master the HuggingFace Hub as your primary interface for finding, evaluating, and deploying open-source models. Learn to read model cards, use the Hub API, and navigate 800k+ models efficiently.

Design rigorous human evaluation studies for LLMs - from annotation protocols to inter-annotator agreement to Chatbot Arena methodology.

How combining attention and Mamba layers creates models that outperform pure architectures - Jamba's design, the attention-to-Mamba ratio, MoE integration, and the emerging hybrid landscape.

Combine BM25 sparse retrieval with dense vector search for best-of-both-worlds performance - understand SPLADE, fusion methods, and when hybrid beats pure dense.

Industrial software development across chip design, GPU optimization, and embedded systems lacks expert reasoning traces showing how engineers reason ab...

Learn how to systematically reduce LLM inference costs using model selection, quantization, caching, request routing, prompt compression, and infrastructure strategies.

Production techniques for serving MoE models efficiently - expert caching, CPU offloading, vLLM support, tensor vs. expert parallelism, batch size sensitivity, and quantization strategies.

Reducing the hardware footprint of large language models (LLMs) during decoding is critical for efficient long-sequence generation. A key bottleneck is...

How instruction tuning transforms base LLMs into general-purpose assistants that can follow diverse instructions, reason step by step, and generalize to new tasks.

A complete guide to Jason Liu's Instructor library - Pydantic-based structured extraction, automatic retry on validation failure, multi-provider support, streaming, and production extraction patterns.

Supervised Semantic Differential (SSD) is a mixed quantitative-interpretive method that models how text meaning varies with continuous individual-differ...

Error Span Detection (ESD) is a crucial subtask in Machine Translation (MT) evaluation, aiming to identify the location and severity of translation erro...

Training Transformer language models is expensive, as performance typically improves with increasing dataset size and computational budget. Although sca...

How safety training gets bypassed - jailbreak taxonomy, GCG attacks, many-shot jailbreaking, prompt injection, defenses, and why the arms race is hard to win.

A complete guide to native JSON mode, OpenAI Structured Outputs, tool calling for structured data, Anthropic tool use, parallel tool calls, and schema design best practices.

Adapter-based methods have become a cost-effective approach to continual learning (CL) for Large Language Models (LLMs), by sequentially learning a low-...

This paper describes the KCLarity team's participation in CLARITY, a shared task at SemEval 2026 on classifying ambiguity and evasion techniques in poli...

Learn how the key-value cache eliminates redundant attention computation in LLM inference, and how PagedAttention solves the memory fragmentation problem.

A thorough guide to LangChain's core abstractions, LCEL composable pipelines, LangGraph stateful workflows, LangSmith observability, and when to use LangChain vs direct API calls.

Learn the training objectives that teach LLMs to understand language - causal language modeling, masked language modeling, cross-entropy loss, and perplexity.

Large language models (LLMs) undergo alignment training to avoid harmful behaviors, yet the resulting safeguards remain brittle: jailbreaks routinely by...

A speaker encoder used in multilingual voice cloning should treat the same speaker identically regardless of which script the audio was uttered in. Off-...

How to decompose LLM latency and cost, choose the right optimization strategies, and define SLOs that balance quality, speed, and budget.

Latent reasoning offers a more efficient alternative to explicit reasoning by compressing intermediate reasoning into continuous representations and sub...

How layer normalization and residual connections solve gradient flow in deep transformers and enable training of 100+ layer networks.

Research on developmentally plausible language models has largely focused on English, leaving open questions about multilingual settings. We present a s...

Large language model (LLM) agents require long-term user memory for consistent personalization, but limited context windows hinder tracking evolving pre...

All prior membership inference attacks for fine-tuned language models use hand-crafted heuristics (e.g., loss thresholding, Min-K\%, reference calibrati...

Although most of the automated theorem-proving approaches depend on formal proof systems, informal theorem proving can align better with large language...

Why the quadratic complexity of self-attention creates real production bottlenecks - memory, latency, and cost - and why sparse attention approximations only partially solve the problem.

How weight averaging of fine-tuned models produces better, more robust models than any individual fine-tune - and the task arithmetic framework for composing capabilities.

A deep dive into Meta's LLaMA model family - from LLaMA 1 through LLaMA 3.3 - covering RoPE embeddings, SwiGLU activation, RMSNorm, grouped query attention, and when to choose each variant.

A comprehensive guide to LlamaIndex's data-centric architecture - indices, query engines, workflows, multi-document agents, and how it compares to LangChain for RAG applications.

Design and operate an LLM gateway - unified API, model routing, circuit breakers, budget enforcement, and fallback chains - using LiteLLM and custom routing logic.

Large language models (LLMs) perform increasingly well on biology benchmarks, but it remains unclear whether they uplift novice users -- i.e., enable hu...

The three fundamental LLM product patterns - chat, workflow automation, and autonomous agents - and how to design the production service graph for each.

Use powerful LLMs to automatically evaluate other models - with position bias mitigation, CoT judging, and cost analysis.

A structured, production-grade LLM curriculum - from transformer architecture to alignment and safety. 17 modules covering every layer of the LLM stack.

How Microsoft Guidance and LMQL extend structured generation to full programmatic control - interleaving generation with code, SQL-like constraints, token healing, and when each tool wins over Outlines and Instructor.

Large language models (LLMs) have driven substantial advances in speech language models (SpeechLMs), yielding strong performance in automatic speech rec...

The widespread adoption of reinforcement learning-based alignment highlights the growing importance of reward models. Various benchmarks have been built...

Master LoRA - the parameter-efficient fine-tuning method that adds only 0.3% of parameters to GPT-3 while matching full fine-tuning quality, making LLM fine-tuning feasible on a single GPU.

The empirical finding that LLMs reliably recall information at the beginning and end of long contexts but miss information in the middle, and strategies to mitigate this U-shaped performance degradation.

Although Automatic Speech Recognition (ASR) in Bengali has seen significant progress, processing long-duration audio and performing robust speaker diari...

How Mamba's input-dependent SSM parameters, hardware-aware parallel scan, and selective gating mechanism achieved linear-time sequence modeling competitive with transformers.

A rigorous benchmark comparison: perplexity, throughput, recall tasks, in-context learning, and the fundamental trade-off between compressed state and full context access.

Recent advances in large language models (LLMs) have enabled the large-scale generation of highly fluent and deceptive news-like content. While prior wo...

Large language model agents receive instructions from many sources-system messages, user prompts, tool outputs, and more-each carrying different levels...

Research on classroom interaction has long been divided between large-scale observation and in-depth ethnographic work. We propose a framework mapping t...

Understand how BERT learns bidirectional language representations using masked language modeling, its architecture, and how to fine-tune it for downstream tasks.

Nested embeddings where any prefix of dimensions is informative - training MRL, adaptive retrieval, 10x FLOP reduction, and how OpenAI's text-embedding-3 uses MRL internally.

Recent work on chain-of-thought (CoT) faithfulness reports single aggregate numbers (e.g., DeepSeek-R1 acknowledges hints 39% of the time), implying tha...

Numerous metascience studies and other initiatives have begun to monitor the prevalence of open science practices when it is more important to understan...

Large language model (LLM) agents are fundamentally bottlenecked by finite context windows on long-horizon tasks. As trajectories grow, retaining tool o...

Designing memory systems for LLM agents - from in-context working memory to episodic retrieval, semantic knowledge bases, and procedural memory.

Large Language Models (LLMs) have demonstrated remarkable capability in machine translation on high-resource language pairs, yet their performance on lo...

How to use arcee-ai/mergekit to merge language models with YAML configuration, CPU-compatible layer-by-layer processing, and automated HuggingFace Hub upload.

Large Language Models (LLMs) are increasingly being deployed in multilingual, multicultural settings, yet their reliance on predominantly English-centri...

Mistral 7B's sliding window attention and grouped query attention innovations, and Mixtral 8x7B's Mixture of Experts design - sparse routing, expert selection, and why MoE delivers 70B quality at 13B active parameter cost.

Mistral AI's Mixtral 8x7B architecture - 8 experts with top-2 routing, sliding window attention, multilingual training, performance vs. Llama 2 70B, and serving requirements.

The architecture of sparse MoE models - how expert networks replace dense FFN layers, top-k routing, and how parameter count relates to active compute.

As Large Language Models (LLMs) are increasingly deployed in cross-linguistic contexts, ensuring safety in diverse regulatory and cultural environments...

Multimodal Stance Detection (MSD) is crucial for understanding public discourse, yet effectively fusing text and image, especially with conflicting sign...

Multimodal LLMs can process speech and images, but they cannot hear a speaker's voice or see an object's texture. We show this is not a failure of encod...

Open-source model licenses are not all the same. Learn Apache 2.0, LLaMA Community, RAIL, and custom licenses - what you can and cannot do in production, and how to build a compliance workflow.

When researchers iteratively refine ideas with large language models, do the models preserve fidelity to the original objective? We introduce DriftBench...

Survey the post-RLHF alignment landscape - RLAIF, Constitutional AI, rejection sampling fine-tuning, iterative DPO, process reward models, and the open questions shaping the next generation of aligned models.

Semi-structured documents integrate diverse interleaved data elements (e.g., tables, charts, hierarchical paragraphs) arranged in various and often irre...

A complete guide to the transformer architecture - the foundation of every modern large language model.

Python patterns for building production LLM applications - API integration, streaming, prompt engineering, token management, tool use, and vector search.

Master the art and science of communicating with large language models - from basic zero-shot instructions to automated prompt optimization with DSPy.

Master Retrieval-Augmented Generation - the dominant pattern for grounding LLMs in external knowledge at production scale.

A complete guide to evaluating large language models - from perplexity to production monitoring.

Master the systems and techniques that make large language model inference fast, efficient, and cost-effective at production scale.

Understanding how modern AI systems process images, audio, and text together - from CLIP to diffusion to production pipelines.

Production architecture for AI-powered products - from prototype to reliable, scalable, cost-efficient systems.

How modern LLMs learn to think - test-time compute, chain-of-thought, process reward models, and the architectures behind o1, o3, and DeepSeek-R1.

How sparse MoE models achieve massive capacity at lower compute cost - routing mechanisms, load balancing, Mixtral, and DeepSeek's innovations.

A complete map of State Space Models - from the quadratic attention bottleneck to Mamba's selective recurrence, hybrid architectures, and production deployment.

A complete map of structured generation - from the reliability problem with free-text LLM output to constrained decoding, Outlines, Instructor, JSON mode, and production-grade extraction pipelines.

How to combine multiple fine-tuned language models into a single, more capable model without any additional training.

How modern LLMs handle extremely long inputs - from the fundamental O(n²) attention problem to RoPE scaling, context compression, and production engineering for 128K+ context windows.

A complete guide to AI alignment, RLHF, Constitutional AI, DPO, red teaming, jailbreaks, safety evaluations, and the global regulatory landscape.

A complete guide to embeddings - models, evaluation (MTEB), fine-tuning, Matryoshka embeddings, quantization, multimodal embeddings, and production pipelines.

LLM agents as autonomous systems that reason, plan, and act using tools, memory, and multi-agent coordination.

Adapting MCTS to language model reasoning - selection, expansion, simulation, backpropagation over reasoning steps, AlphaCode 2, Tree-of-Thought, and production trade-offs.

We present a scalable methodology for evaluating language models in multi-turn interactions, using a suite of collaborative games that require effective...

We present MTRAG-UN, a benchmark for exploring open challenges in multi-turn retrieval augmented generation, a popular use of large language models. We...

Building systems where multiple specialized LLM agents collaborate through orchestrator-worker, pipeline, and peer-to-peer patterns using LangGraph and CrewAI.

Knowledge distillation is an effective technique for pre-trained language model compression. However, existing methods only focus on the knowledge distr...

How multi-head attention enables transformers to jointly attend to information from multiple representation subspaces simultaneously.

CLIP, SigLIP, ImageBind, ColPali, and CLAP - embedding images, text, audio, and documents in shared vector spaces for cross-modal search and zero-shot classification.

How open-source vision-language models work - from CLIP vision encoders and projection layers to LLaVA, InternVL2, and LLaMA 3.2 Vision - and how to deploy them for document understanding, OCR, and visual reasoning in production.

How to build retrieval-augmented generation systems that can retrieve and reason over images, PDFs with figures, slides, and mixed-media documents.

Instruction Tuning (IT) has been proven to be an effective approach to unlock the powerful capabilities of large language models (LLMs). Recent studies...

This paper explores the response of Large Language Models (LLMs) to user prompts with different degrees of politeness and impoliteness. The Politeness T...

We introduce NOBLE (Nonlinear lOw-rank Branch for Linear Enhancement), an architectural augmentation that adds nonlinear low-rank branches to transforme...

Build production observability for LLM applications - distributed tracing, quality metrics, cost attribution, prompt versioning, and drift detection using LangSmith, Langfuse, and Helicone.

Surprisal theory links human processing effort to the predictability of an upcoming linguistic unit, but empirical work often leaves the notion of a uni...

Recent works proposed test-time alignment methods that rely on a small aligned model as a proxy that guides the generation of a larger base (unaligned)...

Run, fine-tune, quantize, evaluate, and deploy open source LLMs in production - the complete hands-on guide for engineers who want to own their models.

text-embedding-3, Matryoshka training, Voyage AI, Cohere Embed, cost analysis, batch processing patterns, and when to choose API vs self-hosted embeddings.

What we know about OpenAI's o1 and o3 reasoning models - hidden chain-of-thought, reinforcement learning from process rewards, compute budget tokens, and ARC-AGI results.

This paper presents a systematic benchmark of state-of-the-art multilingual large language models (LLMs) adapted via token pruning - a compression techn...

A complete guide to the Outlines library - Pydantic schema to FSM, regex constraints, JSON schema constraints, vLLM integration, and production deployment patterns with guaranteed output conformance.

Understand perplexity, cross-entropy, bits per byte, and when intrinsic metrics mislead you about model quality.

Microsoft Phi model family - textbook quality data hypothesis, how 1-4B models can match much larger ones on reasoning tasks, and the design principles behind efficient small language models.

How LLM agents handle complex multi-step tasks through plan-and-execute, hierarchical planning, self-reflection, and LangGraph-based workflows.

Constructing computer-aided design (CAD) models is labor-intensive but essential for engineering and manufacturing. Recent advances in Large Language Mo...

Explainable Artificial Intelligence (XAI) seeks to enhance the transparency and accountability of machine learning systems, yet most methods follow a on...

As large language models (LLMs) transition from research prototypes to real-world systems, customization has emerged as a central bottleneck. While text...

How positional encodings inject sequence order information into transformers - from sinusoidal to RoPE.

We investigate how verbal and nonverbal linguistic features, exhibited by speakers and listeners in dialogue, can contribute to predicting the listener'...

Resource-efficient training optimization techniques are becoming increasingly important as the size of large language models (LLMs) continues to grow. I...

The infrastructure, parallelism strategies, memory optimizations, and training data choices required to pretrain large language models on thousands of GPUs.

In this paper, we propose Precision-Informed Semantic Modeling (PRISM), a structured topic modeling framework combining the benefits of rich representat...

The standard post-training recipe for large multimodal models (LMMs) applies supervised fine-tuning (SFT) on curated demonstrations followed by reinforc...

How process reward models provide step-level supervision for reasoning - the Lightman et al. 2023 paper, Math-Shepherd, using PRMs for search, and their limitations.

Build a comprehensive production monitoring stack for LLMs - latency, cost, quality drift, safety, and observability platforms compared.

Build and operate multimodal AI pipelines at production scale - image preprocessing, cost control, VLM hallucination mitigation, caching, security, and observability for vision-language workloads.

Understand how prompt injection attacks work, why they're hard to defend against, and how to build LLM systems that are resistant to manipulation.

Move beyond manual prompt engineering to automated, evaluation-driven optimization - using APE, OPRO, and DSPy to build LLM pipelines that improve themselves.

Building maintainable prompt systems in Python - template engines, versioning, testing prompts, few-shot construction, and prompt injection defense.

Learn how QLoRA combines 4-bit quantization with LoRA to fine-tune 65B parameter models on a single consumer GPU, using NF4 quantization, double quantization, and paged optimizers.

While second language (L2) learners may acquire target syntactic word order, mapping this syntax onto appropriate prosodic structures remains a persiste...

Master LLM quantization techniques - from LLM.int8() to GPTQ and AWQ - to run large models on commodity hardware without unacceptable quality loss.

Alibaba Qwen and DeepSeek architectural innovations - MLA attention, DeepSeekMoE, multi-token prediction, and how Chinese labs are advancing open-source LLM research.

Build rigorous RAG evaluation with RAGAS, TruLens, LLM-as-judge, golden datasets, and production monitoring - measure faithfulness, relevance, and groundedness.

Evaluate RAG systems with precision - the RAG triad, RAGAS framework, golden datasets, and retrieval metrics for production pipelines.

A rigorous cost, latency, and accuracy comparison of retrieval-augmented generation versus long-context stuffing, with decision frameworks for production use cases.

Building LLM agents that interleave reasoning traces and actions in a ReAct loop to solve multi-step tasks with tool grounding.

Learn how to build LLM agents that reason and act by interleaving thought and tool calls - the architectural pattern behind every modern AI assistant.

Engineering breakdown of the ReAct paper (Yao et al., 2022) - the foundation of every AI agent built today. Plain English, production viability rating, implementation notes.

Large Language Models (LLMs) achieve strong performance on many reasoning benchmarks, yet these evaluations typically focus on isolated tasks that diffe...

We propose RecaLLM, a set of reasoning language models post-trained to make effective use of long-context information. In-context retrieval, which ident...

Reinforcement Learning from Verifiable Rewards (RLVR) has emerged as a powerful paradigm for enhancing the complex reasoning capabilities of Large Reaso...

Systematic adversarial evaluation of language models - manual red teaming, automated red teaming with LLMs, failure taxonomies, and building a production red team process.

We study multiteacher knowledge distillation for low resource abstractive summarization from a reliability aware perspective. We introduce EWAD (Entropy...

Large language models (LLMs) have revolutionized Text-to-SQL generation, allowing users to query structured data using natural language with growing eas...

Recent research has shown that filtering massive English web corpora into high-quality subsets significantly improves training efficiency. However, for...

Master the two-stage retrieval-reranking architecture - cross-encoders, ColBERT, LLM-as-reranker, Reciprocal Rank Fusion, and production latency budgets.

From InstructGPT to DPO to ORPO. Read the 7 most important alignment papers in order — understanding how LLMs are made to follow human intent.

From Chain-of-Thought to production agent architectures. Read the 9 most important agent papers in order — with full engineering context between each one.

From CLIP to GPT-4V to Gemini. Read the 9 most important multimodal AI papers in order — understanding how vision and language were unified.

Reinforcement Learning with Verifiable Rewards (RLVR) significantly enhances the reasoning capabilities of Large Language Models. When applied to RLVR,...

Master the approximate nearest neighbor algorithms powering vector search - HNSW, IVF, IVF-PQ, ScaNN, and DiskANN with parameter tuning and recall-latency trade-offs.

Retrieval-augmented generation (RAG) is a common way to ground language models in external documents and up-to-date information. Classical retrieval sys...

Translating natural language to SQL (Text-to-SQL) is a critical challenge in both database research and data analytics applications. Recent efforts have...

Reward models are central to aligning large language models (LLMs) with human preferences. Yet most approaches rely on pointwise reward estimates that o...

A complete technical walkthrough of Reinforcement Learning from Human Feedback - the three-phase pipeline, reward models, PPO, KL penalty, and the limitations that led to newer approaches.

Understand how RLHF aligns LLMs with human preferences through three phases - SFT, reward model training, and PPO - and why it produced InstructGPT's surprising result that smaller aligned models beat larger unaligned ones.

How Rotary Position Embedding encodes relative positions through complex-plane rotations, why ALiBi achieves length extrapolation with linear biases, and why RoPE became the dominant approach for long-context models.

The algorithms that decide which experts process which tokens - linear routing, expert choice, auxiliary load balancing loss, noisy top-k gating, and the Switch Transformer approach.

Knowledge graph question answering (KGQA) is a promising approach for mitigating LLM hallucination by grounding reasoning in structured and verifiable k...

Humans solve problems by executing targeted plans, yet large language models (LLMs) remain unreliable for structured workflow execution. We propose RunA...

Evaluate LLMs for harmful outputs, social bias, hallucination, and jailbreak vulnerability - including red teaming methodology and production monitoring.

Recursive self-improvement is moving from theory to practice: modern systems can critique, revise, and evaluate their own outputs, yet iterative self-mo...

Master the sampling algorithms that control LLM output diversity - from greedy decoding to nucleus sampling - and learn when to use each in production.

Scientific literature is expanding at an unprecedented pace, making it increasingly challenging to efficiently organize and access domain knowledge. A h...

The lack of reasoning capabilities in Vision-Language Models (VLMs) has remained at the forefront of research discourse. We posit that this behavior ste...

Empirical power-law relationships between LLM performance and compute, data, and parameters - from Kaplan (2020) to Chinchilla (2022) and beyond.

How self-attention computes query, key, and value interactions to capture long-range dependencies between tokens.

On-policy distillation (OPD) has become a popular training paradigm in the LLM community. This paradigm selects a larger model as the teacher to provide...

Large Language Models (LLMs) increasingly serve as autonomous reasoning agents in decision support, scientific problem-solving, and multi-agent coordina...

Large language models (LLMs) have demonstrated remarkable capabilities across diverse tasks. However, the truthfulness of their outputs is not guarantee...

We present a dataset and a model for sentiment analysis of German sign language (DGS) fairy tales. First, we perform sentiment analysis for three levels...

How spherical linear interpolation provides smoother, geometrically correct blending between two model weight configurations than simple linear averaging.

Recently, there have been significant advancements in music generation. However, existing models primarily focus on creating modern pop songs, making it...

Why MoE gives more capacity per FLOP than dense models, the memory vs. compute trade-off, training efficiency, inference complexity, and when to choose each architecture.

Real-world Table-Text question answering (QA) tasks require models that can reason across long text and source tables, traversing multiple hops and exec...

Automatic speech recognition (ASR) has benefited from advances in pretrained speech and language models, yet most systems remain constrained to monoling...

Learn how speculative decoding uses a small draft model to generate tokens that a large target model verifies in parallel, achieving 2-3x speedup with no quality loss.

Large Language Models (LLMs) are increasingly used as proxies for human perception in urban analysis, yet it remains unclear whether persona prompting p...

How control theory's state space models became a competitive sequence modeling architecture - continuous-time SSMs, the S4 paper, HiPPO initialization, and the convolutional/recurrent duality.

As AI-generated fiction becomes increasingly prevalent, questions of authorship and originality are becoming central to how written work is evaluated. W...

Streaming LLM output in Python - server-sent events, async generators, FastAPI streaming endpoints, and building real-time chat UIs.

Long conversations with an AI agent create a simple problem for one user: the history is useful, but carrying it verbatim is expensive. We study persona...

Production-grade architecture for structured generation pipelines - reliability stacks, schema versioning, monitoring, async batching, caching, edge case handling, and complete reference implementations.

Reliably extract structured data from LLMs using JSON mode, function calling, Pydantic validation, and constrained decoding - the backbone of production LLM pipelines.

Learn how to adapt pretrained LLMs to specific tasks through supervised fine-tuning - data preparation, hyperparameters, catastrophic forgetting, and evaluation.

Recent advances in language models have substantially improved Natural Language Understanding (NLU). Although widely used benchmarks suggest that Large...

Realistic long-horizon productivity work is strongly conditioned on user-specific computer environments, where much of the work context is stored and or...

Master the architecture of LLM conversations - how to design system prompts, manage context windows, and build production-grade context management systems.

Modern optimizers like Adam and Muon are central to training large language models, but their reliance on first- and second-order momenta introduces sig...

Large language models (LLMs) with reasoning capabilities have fueled a compelling narrative that reasoning universally improves performance across langu...

Small language models (SLMs) have emerged as efficient alternatives to large language models for task-specific applications. However, they are often emp...

Traditional vision-language models struggle with contrastive fine-grained taxonomic reasoning, particularly when distinguishing between visually similar...

Learn how tensor parallelism splits weight matrices across GPUs and pipeline parallelism splits model layers, enabling inference and training of models too large for a single GPU.

This study presents the first systematic, reference-free human evaluation of large language model (LLM) machine translation (MT) for Ancient Greek (AG)...

The paradigm shift from training-time scaling to inference-time scaling - best-of-N sampling, majority voting, and how spending more compute at inference improves reasoning quality.

Why making AI systems do what we actually want is harder than it looks - the specification problem, Goodhart's Law, reward hacking, and outer vs inner alignment.

This study explores artificial visual creativity, focusing on ChatGPT's ability to generate new images intentionally pastiching original artworks such a...

Why attention is O(n²) in memory and compute, how the KV cache grows with context length, and how FlashAttention solves the IO bottleneck without changing the algorithm.

Large Language Models (LLMs) often exhibit highly agreeable and reinforcing conversational styles, also known as AI-sycophancy. Although this behavior i...

Personal Artificial Intelligence is currently hindered by the fragmentation of user data across isolated silos. While Retrieval-Augmented Generation off...

How TIES-Merging eliminates task interference by trimming small deltas, electing signs by majority vote, and merging only aligned parameters.

Tiktoken, tokenisation internals, context window management, sliding window strategies, and building cost-aware LLM applications.

How tokenizers convert raw text to token IDs - BPE from scratch, WordPiece, byte-level BPE, and the surprising ways tokenization breaks models.

Enabling LLMs to invoke external tools and APIs through structured function calling, covering JSON schema design, Anthropic vs OpenAI formats, parallel tool calls, and production safety.

Building LLM tool use systems in Python -- function calling, tool schemas, execution loops, error handling, and multi-step agent patterns.

Large Language Models (LLMs) have advanced Table Question Answering, where most queries can be answered by extracting information or simple aggregation....

Case Report Forms (CRFs) collect data about patients and are at the core of well-established practices to conduct research in clinical settings. With th...

Vision-language models (VLMs) show promise in drafting radiology reports, yet they frequently suffer from logical inconsistencies, generating diagnostic...

How to train Mixture of Experts models at scale - expert parallelism, capacity factors, token dropping, load imbalance, training instability, and the GShard approach to 600B parameters.

Mathematical text understanding is a challenging task due to the presence of specialized entities and complex relationships between them. This study for...

Explore multiple reasoning paths simultaneously using Tree-of-Thought - the technique that enables LLMs to backtrack, evaluate alternatives, and solve problems that defeat linear chain-of-thought.

UI-to-Code generation requires vision-language models (VLMs) to produce thousands of tokens of structured HTML/CSS from a single screenshot, making visu...

Despite their capabilities, Multimodal Large Language Models (MLLMs) may produce plausible but erroneous outputs, hindering reliable deployment. Accurat...

Reinforcement Learning with Verifiable Rewards (RLVR) has achieved substantial gains in single-attempt accuracy (Pass@1) on reasoning tasks, yet often s...

Cooking is a cultural expression of human creativity that transcends geography and time through the orchestration of ingredients and techniques, much li...

We present a method to identify a valence-arousal (VA) subspace within large language model representations. From 211k emotion-labeled texts, we derive...

Compare Pinecone, Qdrant, Weaviate, Milvus, Chroma, and pgvector - understand the engineering trade-offs and build a production vector store.

Video agentic models have advanced challenging video-language tasks. However, most agentic approaches still heavily rely on greedy parsing over densely...

How modern AI systems combine vision encoders with language models to understand and reason about images.

Vision-language models (VLMs) still struggle with visual perception tasks such as spatial understanding and viewpoint recognition. One plausible contrib...

Large Vision Language Models (LVLMs) achieve strong multimodal reasoning but frequently exhibit hallucinations and incorrect responses with high certain...

Learn how production inference servers like vLLM, TGI, TensorRT-LLM, and Ollama combine PagedAttention, continuous batching, and optimized kernels to serve LLMs at scale.

The fundamental concept of embeddings - mapping meaning to geometric space, cosine similarity, Word2Vec, the king-queen analogy, and why dense retrieval replaced keyword search.

We investigate the separation of literal interpretation from contextual inference in a collaborative block-building task where a builder must resolve un...

As Large Language Models (LLMs) increasingly mediate global information access with the potential to shape public discourse, their alignment with univer...

Large language models (LLMs) often achieve strong performance on reasoning benchmarks, but final-answer accuracy alone does not show whether they faithf...

Background: Patient-facing medical chatbots based on retrieval-augmented generation (RAG) are increasingly promoted to deliver accessible, grounded heal...

A practical decision framework for routing tasks to reasoning models - task taxonomy, cost-benefit analysis, latency trade-offs, and hybrid routing architectures.

A practical deployment guide: use cases where SSMs win, the streaming inference pattern, model availability on HuggingFace, fine-tuning SSMs, and a forward-looking outlook.

Diffusion Language Models (DLMs) are often advertised as enabling parallel token generation, yet practical fast DLMs frequently converge to left-to-righ...

The catastrophic forgetting problem, why naive ensembles are too expensive, and the surprising geometric insight that makes model merging possible.

Understand why LLMs hallucinate, what RAG actually solves, and the decision framework for choosing RAG vs fine-tuning vs prompt stuffing.

The taxonomy of LLM output failures, why prompt-based JSON extraction breaks at scale, the production impact of 5% failure rates, and the spectrum of solutions from prompt engineering to constrained decoding.

A complete production engineering guide for building applications with long-context LLMs - model selection, cost management, prompt structure, multi-turn conversation, and memory-augmented systems.

Recent work interprets the linear recoverability of geographic and temporal variables from large language model (LLM) hidden states as evidence for worl...

Norm, the formal theoretical linguist, and Claudette, the computational language scientist, have a lovely time discussing whether modern language models...

Learn how to elicit reliable behavior from LLMs using only instructions - no examples required - by mastering prompt anatomy, role personas, and format control.