Knowledge catalogue
arXiv:2608.15284v1 Announce Type: cross Abstract: Navigation instruction generation from ego-centric RGB video in continuous environments is an important yet challenging task for human-robot interacti
arXiv:2608.16045v1 Announce Type: cross Abstract: LLM-based data-analysis tools are increasingly used to help users analyze messy spreadsheets and workbooks, from answering questions over uploaded fil
arXiv:2608.14747v1 Announce Type: new Abstract: WANDR (Wide ANd Deep Research) is a benchmark of 500 realistic, challenging data-collection tasks for research agents. Each task requires a system to di
arXiv:2608.14573v1 Announce Type: cross Abstract: Large language model (LLM) agents are increasingly capable of tool use, code execution, artifact inspection, and iterative revision, creating new oppo
arXiv:2505.04608v5 Announce Type: replace-cross Abstract: Responsibly deploying artificial intelligence (AI) / machine learning (ML) systems in high-stakes settings arguably requires not only proof of
We have paused some frontier RL training to ensure that we can meet the appropriate alignment, security and monitoring standards for the new level of capabilities in front of us. Model progress is now
We just released TensorRT Model Connect in Public Preview. You can take a supported @huggingface model to end-to-end TensorRT inference in just two commands. No intermediate ONNX export, and the resul
AI companies like Anthropic and OpenAI regularly publish reports on how people are using products like Claude and ChatGPT, but they only release the data they want us to see, AI researchers say. “Ther
We’re sharing the concrete changes we’re making to strengthen monitoring, security, and alignment as capabilities advance. We’ve introduced stronger workload and network isolation, continuous security
arXiv:2608.15092v1 Announce Type: cross Abstract: In this work, we introduce WeSCE, a benchmark for quantifying security drift in code editing under weak-security constraints, where tasks specify only
we've been doing a lot of a/b testing of @aiDotEngineer youtube thumbnails. i always hated that it is such an opaque process. open sourcing/crowdsourcing our learnings today! https://ai-engineer-thumb
What can GPT 5.6 sol do that GPT 5.5 xhigh couldn’t ?? As always @GaryMarcus vindicated This isn’t a fucking issue about models getting “too smart.” Quite the opposite. ChatGPT is delaying because the
arXiv:2608.16852v1 Announce Type: new Abstract: Regulatory compliance monitoring in deployed language models is increasingly implemented as a legal and audit control, checking model outputs against wr
arXiv:2608.16370v1 Announce Type: new Abstract: Task completion is the standard metric for evaluating context compression, yet it is incomplete: compression can increase an agent's interaction cost by
arXiv:2608.14819v1 Announce Type: cross Abstract: Music foundation models are commonly used as frozen audio feature extractors, yet selecting which layer to extract from remains largely heuristic. Cur
Hi everyone, I use AI agents to code and manage my personal files heavily. However, I am worried that the agents may accidentally delete some important files on my machine, outside of the defined scop
arXiv:2608.14841v1 Announce Type: new Abstract: Long-document visual question answering (VQA) over documents of tens to hundreds of pages mixing text, tables, charts, and figures typically follows ret
arXiv:2608.14855v1 Announce Type: new Abstract: Machine unlearning aims to remove targeted data or behaviors from a trained model without retraining from scratch. Yet most evaluations assume that the
arXiv:2608.15708v1 Announce Type: new Abstract: We study a critical yet overlooked failure mode in Grounded Video Question Answering: question-invariant grounding, where models predict nearly identica
Theoretical use case: I have a compressed version of wikipedia, and some function or tool fooRAG. Then I set my system prompt (or make a skill, or hook, etc.) to include 'You can search wikipedia with
arXiv:2608.14680v1 Announce Type: new Abstract: Reliability in LLM-based agentic systems is a property of the whole execution (its tool calls, model calls, guardrails, and inter-agent messages), not o
arXiv:2608.16801v1 Announce Type: new Abstract: We study how teams of AI coding agents coordinate while solving programming tasks. Current evaluations usually report whether the agents complete the ta
arXiv:2608.15338v1 Announce Type: cross Abstract: Sentiment classifiers are increasingly applied to social media content that is either sarcastic or AI-generated --- two distributional regimes where s
arXiv:2608.16515v1 Announce Type: cross Abstract: Retrieval-augmented generation (RAG) improves large language models by grounding generation in external evidence, but it also introduces a source trus
arXiv:2608.15323v1 Announce Type: new Abstract: Understanding a model and its learning mechanisms in depth requires identifying when its internal structures become useful, rather than simply looking a
arXiv:2608.16627v1 Announce Type: cross Abstract: Natural language explanations (NLEs) are increasingly used as inputs, for example, as few-shot rationales that influence model behavior in in-context
arXiv:2608.14610v1 Announce Type: new Abstract: Legal reasoning tasks such as legal judgment prediction (LJP) require identifying the temporally correct version of the law governing a case -- a capabi
arXiv:2608.14875v1 Announce Type: cross Abstract: Machine-learned exchange-correlation functionals correct band gaps at near-semilocal cost, while density-functional tight binding reaches the 10^3-10^
arXiv:2608.14665v1 Announce Type: new Abstract: The temperature that maximizes pass@k is often low for a small sampling budget and higher for a large budget. This pattern has been reported from Codex
arXiv:2608.15592v1 Announce Type: new Abstract: Efficient LLM serving is often bottlenecked by the need to pad sequences to a fixed maximum length, and this wastes compute and degrades throughput. Pre
arXiv:2608.14940v1 Announce Type: new Abstract: Current agent evaluations score models on the state visible at the end of a stopped run which they count as one trial. However, interpreting the score a
arXiv:2608.15639v1 Announce Type: cross Abstract: extit{Split Federated Learning} (SFL) enables distributed model training by splitting networks between the server and clients. However, under client h
arXiv:2608.15879v1 Announce Type: new Abstract: Large language models (LLMs) have shown strong performance in text generation tasks, yet their effectiveness on headline generation remains sensitive to
arXiv:2608.16147v1 Announce Type: new Abstract: Class-imbalance handling is routinely evaluated on a single benchmark dataset, and the resulting conclusions are reported as if they were properties of
arXiv:2608.16806v1 Announce Type: cross Abstract: Large Language Models (LLMs) have demonstrated capabilities in in-context learning, task decomposition, step-by-step reasoning, and code generation, d
arXiv:2608.15654v1 Announce Type: cross Abstract: Large language models can write fluent stories, but open-ended storytelling requires more than local fluency. In evolving world simulations and AI-nat
arXiv:2608.14559v1 Announce Type: new Abstract: Effective communication in multi-agent reinforcement learning requires agents to decide not only extit{what} to communicate, but when? Existing approach
arXiv:2608.16502v1 Announce Type: new Abstract: Large-scale agents increasingly rely on retrieval to access external capabilities. We study this retrieval gate in structured tools and APIs, a measurab
arXiv:2608.14659v1 Announce Type: new Abstract: Large language models for code generation often produce incorrect solutions without reliable indicators of failure. We study whether uncertainty estimat
arXiv:2608.14886v1 Announce Type: new Abstract: Retrieval quality in RAG systems is commonly reported as a single aggregate score, which can hide large differences across query types and language cond
arXiv:2608.14811v1 Announce Type: new Abstract: Teams that adopt cycle-consistent adversarial networks for unpaired image-to-image translation meet the same obstacles: adversarial training oscillates
arXiv:2608.14712v1 Announce Type: cross Abstract: Each row of a transformer's attention matrix is a probability distribution over tokens, and in trained models most of that probability lands on a sing
arXiv:2608.15510v1 Announce Type: new Abstract: Chart-to-code generation requires a model to read the fine-grained visual details of a chart and write executable code that reproduces it. Existing char
arXiv:2608.15980v1 Announce Type: new Abstract: Preference benchmarks are built by hiring annotators, and the identity of those annotators is treated as an implementation detail. We measure what that
arXiv:2608.15530v1 Announce Type: new Abstract: Reinforcement learning with human feedback (RLHF) aligns LLMs with human preferences, improving summarization fluency and safety, but causes sentiment d
arXiv:2608.15085v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) exhibit substantial performance degradation in non-English visual reasoning, despite the strong multilingual co
arXiv:2608.15641v1 Announce Type: new Abstract: This paper evaluates Wiktionary as an ethically crowdsourced lexicon for English dialects. We took a two-phase approach, providing an in-depth descripti
arXiv:2608.14604v1 Announce Type: cross Abstract: Small language models in the ten to one hundred million parameter range are attractive for on device inference, rapid experimentation, and controlled
arXiv:2608.14633v1 Announce Type: cross Abstract: Wolff-Parkinson-White (WPW) syndrome is a congenital cardiac pre-excitation, clinically important and often missed on the resting 12-lead ECG. Detecti
arXiv:2608.14876v1 Announce Type: cross Abstract: Agentic coding assistants are finding widespread use, not just in new code development but in quickly ingesting and leveraging third-party code. This
arXiv:2510.18135v2 Announce Type: replace Abstract: Generative world models (WMs) can now simulate worlds with striking visual realism, which naturally raises the question of whether they can endow em
arXiv:2608.15659v1 Announce Type: new Abstract: Learning to generate or reconstruct explorable worlds requires video paired with more than RGB: camera motion, scene geometry, temporal correspondence a
arXiv:2608.16747v1 Announce Type: cross Abstract: Many areas of AI research, such as language model interpretability and chain of thought faithfulness, seek to explain model behaviors. But what consti
😮 wow that’s huge 💰 🚀Today we launched LangSmith Tuned Evaluators, starting with Perceived Error. Tuned Evaluators run on production traces to catch undesirable agent behavior and attach feedback that
arXiv:2608.14843v1 Announce Type: cross Abstract: As authorship attribution systems are increasingly deployed to detect ghostwritten and AI-generated papers, their errors can support accusations again
arXiv:2608.16658v1 Announce Type: cross Abstract: Cross-view Video Geo-localization (CVG) aims to localize ground-view videos by retrieving their corresponding geo-tagged aerial images. However, CVG a
arXiv:2608.14700v1 Announce Type: new Abstract: Precise emotion control in audio-driven talking heads remains a challenge due to the reliance on implicit emotion regulation in existing systems, which
arXiv:2608.15713v1 Announce Type: new Abstract: Automated pavement-distress detection is commonly framed as a small-object problem, motivating high-resolution P2/4 detection heads and lossless downsam
arXiv:2608.14854v1 Announce Type: new Abstract: While Multimodal Large Language Models (MLLMs) excel in general video understanding, their capability in fine-grained and motion-centric tasks remains l
arXiv:2608.14796v1 Announce Type: new Abstract: Prostate cancer claims a life every 80 seconds. Early detection is needed to prevent disease progression, and both PSA density calculation and biopsy de