New
Everything new in AI as it appears: launches announced by the labs, brand-new models, datasets, papers and open-source tools. Checked every minute or two. Launches feed · Trending · News
Live now
- Connecting…
Thursday 15 October
- PaperExplore Once, Evaluate Many: Multiple Policy Evaluation for Stochastic Linear Bandits
We study how to evaluate $K$ target policies simultaneously in stochastic contextual bandits with linear reward structure. Given a known $d$-dimensional feature map and r
- PaperNot X, but When: Diagnosing a Mandarin LLM Writing Tic through Contextual Appropriateness
Recurring expressions in model-written text are often treated as signatures of machine authorship, yet most of them serve legitimate functions. Whether such an expression
- PaperDisagreement-Guided Weak-to-Strong Routing
Small language models are increasingly deployed on edge devices and robotic systems to meet cost and latency constraints, yet their limited capabilities can make reliable
- PaperLanguage Is a Security Surface for Agent Guards
Step-level guards decide, before each tool call, whether an agent's next action is safe given the user's request and the trajectory so far. Users write those requests in
- PaperShortcut Sensitivity in Agent-Safety Evaluations
AI agents take actions through tools. Some actions are sensitive, so guards that supervise them must themselves be evaluated reliably. Yet the data used to evaluate a gua
- PaperSurvICL: a Foundation Model for in-context Survival Analysis
Tabular foundation models (TFMs) now outperform gradient-boosted trees in classification and regression. Yet in survival analysis, which predicts when an event occurs fro
- PaperMMSearch-OPD: Easy Made Hard, Hard Made Learnable for Long-Horizon Multimodal Search Agents
Multimodal deep search is moving from short queries to long-horizon investigations that continually acquire evidence. However, training such agents remains constrained by
- PaperQuery-Efficient Rejection Sampling for Parallel Masked Diffusion
Masked diffusion models (MDMs) accelerate generation by unmasking multiple tokens in parallel, but sampling these tokens independently ignores their correlations and bias
- PaperTraining-Free On-Policy Distillation
On-policy distillation consolidates specialist models into one student at the price of rollouts, teacher passes and gradient steps for every teacher. We show that the con
- PaperOS-Helix: Unifying GUI Interaction and Code Execution in Native Computer-Using Agents
Real desktop automation is neither purely visual nor purely programmatic. Agents must understand user-facing interfaces while efficiently manipulating files, scripts, and
- PaperWaveLLM: Wavelet-Domain LLM Entropy Modeling for Lossless Image Compression
Lossless image compression fundamentally relies on accurate entropy modeling, driving a recent shift toward large language model (LLM)-based methods. However, existing ap
- PaperArchitecting for Hardware Diversity: Meta's End-to-End Co-Design for High-Performance AI Training on AMD GPUs
Meta deploys a heterogeneous fleet comprising NVIDIA GPUs, AMD GPUs, and proprietary MTIA accelerators to reduce costs and mitigate vendor lock-in. However, the extensive
- PaperHow Much Physics Can Vision Learn from Audio? Probing and Distilling Damping Structure
Physical understanding is crucial for decision making, yet vision models struggle to infer properties such as damping that govern how materials behave. Audio encodes such
- PaperSkill-Decoupled On-Policy Expert Distillation for Unified Vision-Language Models
Reinforcement learning (RL) can elicit specialized visual capabilities in vision-language models (VLMs), yet jointly optimizing heterogeneous objectives often introduces
- PaperFutureLift: Lifting Trajectory Proposals into Bidirectional Future Graphs for Motion Forecasting
Accurate motion prediction is critical for safe autonomous driving, yet even strong forecasters leave residual errors in their multimodal proposals. Existing refiners com
- PaperGNN-Guided Local Refinement for Efficient Layout Decomposition with DSA-MP
Multiple patterning lithography (MPL) has been widely used to improve lithography resolution in advanced nodes, but it faces cost limits as feature sizes shrink. Directed
- PaperGeneralization Guarantees of Randomized Block Coordinate Descent for Nonconvex Learning
Randomized block coordinate descent (RBCD) is a popular optimization algorithm for many large-scale machine learning problems. The existing theoretical analysis of RBCD m
- PaperTRACE: Topology-Aware Reconnection via Prediction-Adaptive Connectivity Enforcement for Thin-Structure Segmentation
Thin, elongated structures are particularly sensitive to connectivity errors, as even a few local segmentation mistakes can disrupt otherwise continuous paths. Existing s
- PaperBeyond Downstream QA: Rethinking How We Evaluate Agent Memory
Agent memory benchmarks evaluate whether an agent's responses reflect correct knowledge, measuring downstream question-answering accuracy as a proxy for memory quality. W
- PaperDecentralized LLM Workflows via Sequential Auctions
We present a scalable framework for solving multi-step tasks through a decentralized market of self-interested agents. Prior theoretical work demonstrated that agents org
- PaperOpenGCUA: Toward a Fully Open Stack for Generalist Computer Use
Computer use has reached a turning point: the focus is expanding from GUI-only interaction to systems that complete tasks on users' computers through visual and programma
- PaperFunctional Welfare for AI Safety
AI welfare is generally treated purely as an ethical issue rather than also as a safety issue. We argue this is a mistake. Whatever their moral status, models exhibit wel
- PaperHy-Memory-30B-A3B: Attributed Reinforcement Learning for Long-Term Memory Writing
External-memory systems extend language-model applications beyond a context window, but their quality depends critically on the model executing the write path: deciding w
- PaperHTN-Orchestrate: A Hierarchical Task Network Framework for Multi-Agent Orchestration
Orchestrating multiple agents on a complex task requires reconciling local subtask execution with global task planning. Dynamic dispatch approaches provide execution flex
- PaperAutoSFT: An Agentic System for Automated Data Research in Supervised Fine-Tuning
Language model agents increasingly automate post-training, but when given the full training stack they converge on supervised fine-tuning and spend most effort revising t
- PaperTerminal-RSI: Recursive Self-Improving Terminal Agents via Recursive Self-Rewrite and Policy-Refine
Converting verifier outcomes into useful policy updates and partial progress into complete task resolution is crucial for terminal agents. We present Terminal-RSI, which
- PaperAttention over noisy examples: generalization and memorization in the exponential Hopfield network
In Hebbian learning from examples, a Hopfield network retrieves archetypes seen only through noisy examples. We study it in the exponential Hopfield model: one softmax at
- PaperChange the Number, Change the Answer: Assessing Rating Scale Sensitivity in LLM-Based Text Scoring
Large language models (LLMs) are increasingly used in computational social science to assign scalar scores to social and psychological constructs. However, studies use di
- PaperEVA-Cap: Optimizing Audiovisual Video Captioning via Event-Centric Alignment
Recent advances in Omni Language Models (OLMs) provide a unified solution for audiovisual video captioning. Nevertheless, existing approaches predominantly treat captioni
- PaperWhole-Body Dexterity
Current humanoid dexterous manipulation remains largely confined to end-effectors and simple grasps, falling short of humans' ability to engage the entire body by bracing
- PaperErase the Mark, Preserve the Motion: Temporally Consistent Removal of Video Watermarks
Recent advances in video generation have increased the demand for reliable watermarking for copyright protection, yet the robustness of video watermarks against removal a
- PaperLAMP: Layerwise Adaptive Mixing of Predictors for Training-Free Few-Shot Vision-Language Adaptation
Few-shot adaptation often relies on an encoder's final global representation, leaving complementary information in intermediate layers underused. Combining these features
- PaperLCT: Losslessly Compressed Tensors with GPU friendly execution for Training
Lossless tensor compression could reduce memory usage during training, but training requires tensors to be repeatedly re-encoded. We introduce Losslessly Compressed Tenso
- PaperREKI: Registering Knowledge in VLAs
We introduce REKI, a 2.4B vision-language-action model (VLA) pretrained exclusively on public data that sets a new Pareto frontier between performance and model size: on
- PaperMotion- and Gravity-Aware Feedforward 3D Reconstruction
Feed-forward transformer models have recently achieved strong performance in multi-view 3D reconstruction, but they still rely primarily on appearance-based attention and
- PaperMosaic: Multi-Session Agentic Reinforcement Learning
Most existing agentic reinforcement learning methods assume single-session trajectories, where interaction proceeds within one continuously accumulated context. However,
- PaperHigh prevalence of antagonizing rhetoric in trending topics on Twitter/X
Disagreement between political opponents is a central element of democracy. This disagreement can be voiced in different ways. An opponent can be antagonized, i.e. portra
- PaperLexical Reference Frames: Interpreting LM Residual Streams in Native Semantics
The vocabulary of a language model is its native semantics, and the embedding and unembedding fix the meaning of a state at the two ends of the residual stream. Between t
- PaperCLUE: Uncovering Social Context for Pragmatic Understanding
Understanding a conversation requires connecting what is said to the speakers' relationships, expectations, and earlier exchanges. Prior work shows that supplying relevan
- PaperWhere Prefill Expert Reduction Hurts MoE Models: A Few Prompt Tokens Carry Most of the Short-Answer F1 Loss
Running prompt tokens through fewer routed experts during prefill is a simple way to make mixture-of-experts (MoE) serving cheaper, yet it is unclear what this damages. W
Headlines and links come from the labs' own blogs, Hacker News and the GDELT news index, and from the sources our index already follows. We show a short excerpt and always link to the original. A launch reported by a news site rather than the lab is marked “Reported” until the lab itself announces it.