Blog Promote Sign in Sign up

Catalogs

Papers

3,241 in the directory.

Browse AI research papers and methods. Search paper profiles, read recorded summaries and follow source links for the full research.

A–Z Recently synced
Explore other catalogs
Models Datasets Demos Devices Robots MCPs Papers Investors Fundraises Repositories News
How to choose papers

Use the summary to find relevant work, then read the original paper. Compare the method, evaluation setting, baselines and limitations with your own problem. Check whether code and datasets are available; a reported benchmark result may not transfer to your use case.

Understand our sources and data limits

  1. The Handoff Tax: Continuing Non-Native Trajectories in LLM Agents taaft Coding agents perform long-running tasks spanning dozens of model calls, tool uses, and code edits.… 2026-08-25 published
  2. The History Is the Detector: Executing CVE Patch History, End-to-End taaft Public vulnerability databases collect rich information about known software flaws, including their… 2026-09-04 published
  3. The Illusion of $\textit{What If}$: Evaluating the Breakdown of Counterfactual Reasoning in LLMs taaft Counterfactual reasoning requires models to reason beyond the observed world and explain how altere… 2026-08-28 published
  4. The Illusion of Equivalency: Statistical Characterization of Quantization Effects in LLMs taaft Post-Training Quantization has become widely used to compress large language models to make them de… 2026-07-09 published
  5. The Impact of Synthetic Data Augmentation on Discourse-Pragmatic Function Classification taaft Synthetic data augmentation has become a common strategy for addressing class imbalance in NLP, but… 2026-09-03 published
  6. The Implications of Linguistic Illegibility for LLM Security taaft LLMs are trained to generate natural language. However, various strands of evidence indicate that a… 2026-09-02 published
  7. The Interlingua Hypothesis: LLMs Translate via a Latent Task-agnostic Feature Space taaft Large language models (LLMs) have recently demonstrated improved machine translation performance ov… 2026-09-01 published
  8. The Intervention Gap in Latent World Models taaft Planning-time intervention fidelity is a distinct, measurable property of a learned world model: wh… 2026-08-30 published
  9. The Invisible Editorial Layer: Formalizing Undisclosed Inference-Time Steering, Probability Placement, and the Attribution Problem in Deployed Language Models taaft Evaluations of generative language models frequently interpret observable behavioral traits, such a… 2026-08-25 published
  10. The Irreversibility Budget: Fleet-Level Risk Accounting and Admission Control for Agent Operating Systems taaft Fleets of LLM agents now externalize effects that cannot be fully undone: they move money, deploy c… 2026-08-31 published
  11. The Language of the Question Selects the Market: Query Language and Exit IP as Separable Factors in Commercial Recommendations from a Generative Search Interface taaft When a generative search interface answers a commercial question, which market's products it names… 2026-08-30 published
  12. The Memory Trust Gap: Capability-Dependent Failures in Persistent-Memory Agents taaft Persistent memory supports personalized agents, but a stale stored fact can override current author… 2026-09-01 published
  13. The Mirage of Optimizing Training Policies: Monotonic Inference Policies as the Real Objective for LLM Reinforcement Learning taaft Reinforcement learning (RL) has gained growing attention in large language model (LLM) post-trainin… 2026-06-28 published
  14. The Mirror Agent Model: a Bayesian Architecture for Interpretable Agent Behavior taaft In this paper we illustrate a novel architecture generating interpretable behavior and explanations… 2026-09-04 published
  15. The Missing Temporal Link: Temporal Context Routing for Script-Driven Audio-Video Generation taaft Joint audio-video generation models have made substantial progress in visual quality and audio-visu… 2026-09-02 published
  16. The Multiple Timescales of Gradient Descent on the Edge of Stability: A Perturbative Derivation of the Central Flow taaft The central flow of Cohen et al. (2025) is an empirically accurate continuous-time model of gradien… 2026-09-01 published
  17. The Neuro-Symbolic Concept Learner: Interpreting Scenes, Words, and Sentences From Natural Supervision IBM We propose the Neuro-Symbolic Concept Learner (NS-CL), a model that learns visual concepts, words,… 2019-04-26 published
  18. The Pile: An 800GB Dataset of Diverse Text for Language Modeling taaft Recent work has demonstrated that increased training dataset diversity improves general cross-domai… 2020-12-31 published
  19. The Platonic Representation Hypothesis taaft We argue that representations in AI models, particularly deep networks, are converging. First, we s… 2024-05-13 published
  20. The potential of virtual triage AI to improve early detection, care acuity alignment, and emergent care referral of life-threatening conditions Infermedica Asks how often people whose symptoms warrant emergency care arrive at virtual triage with no intent… Infermedica
  21. The Price of Intelligence: A Quality-Adjusted Price Index for AI Services taaft Posted prices for AI inference have fallen steadily since 2024, yet the measured speed of that fall… 2026-08-30 published
  22. The Privacy-Hallucination Tradeoff in Differentially Private Language Models taaft Both privacy and factual accuracy are paramount in high-stakes domains like healthcare. Concerningl… 2026-08-31 published
  23. The Psychological Costs of Artificial Intelligence Adoption in Software Engineering taaft Artificial intelligence (AI) is increasingly used to augment software engineering (SE) workflows. W… 2026-09-03 published
  24. The quality of diagnosis and triage advice provided by free online symptom checkers and apps in Australia Infermedica Benchmarks 36 free website and app symptom checkers available in Australia against 48 medical condi… Infermedica
  25. The Reasoning Error About Reasoning: Why Different Types of Reasoning Require Different Representational Structures taaft Different types of reasoning impose different structural demands on representational systems, yet n… 2026-03-23 published
  26. The Red Queen Gödel Machine: Co-Evolving Agents and Their Evaluators taaft Self-improving agents are state-of-the-art (SOTA) on agentic coding benchmarks and have recently be… 2026-06-24 published
  27. The Rise and Potential of Large Language Model Based Agents: A Survey taaft For a long time, humanity has pursued artificial intelligence (AI) equivalent to or surpassing the… 2023-09-14 published
  28. The Rise of Verbal Reinforcement Learning taaft Natural language is emerging as a primary feedback channel for improving language agents, capable o… 2026-09-01 published
  29. The Safeguard Worked. Is the LLM System Safer? taaft Safeguards in deployed LLM services are evaluated by refusal, attack success, and policy violation… 2026-09-01 published
  30. The Shadow Price of Intelligence: Quality Degradation in LLM Inference as a Supply Chain Problem taaft Large language model providers are compute constrained, and their universal response to congestion… 2026-08-25 published
  31. The Shape of Time: Video-Token Contrast for Temporal Understanding in VideoLMs taaft Seeing frames in order does not mean representing time. Modern VideoLMs receive ordered video strea… 2026-09-03 published
  32. The Stack: 3 TB of permissively licensed source code taaft Large Language Models (LLMs) play an ever-increasing role in the field of Artificial Intelligence (… 2022-11-20 published
  33. The Structure of Quantization Damage in LLMs: Why the Next Bit Should Be Spent Globally taaft Post-training quantization (PTQ) is widely used to reduce the cost of serving large language models… 2026-09-01 published
  34. The Value of Human Expertise taaft We consider optimization applications with unknown parameters where the decision maker believes tha… 2026-08-26 published
  35. The Visual Insensitivity Gap: Diagnosing When Vision-Language Models Fail to Use Visual Evidence taaft Vision-language models are evaluated by aggregate accuracy on multimodal benchmarks, a practice tha… 2026-09-01 published
  36. The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning taaft The White House Executive Order on Artificial Intelligence highlights the risks of large language m… 2024-03-05 published