Tao Gui

This work introduces ReDraft (Reference-Driven Revision and Fine-Tuning), which obtains both from the model's own failures: using an expert response only as a reference, it has the model revise its own incorrect rollout, keeps the revision only if a verifier accepts it, and fine-tunes on what survives.

Atria Dawn Preview is introduced, a foundation agentic language model designed for scientific research and engineering workflows, with the goal of expanding the frontier of agent productivity in the real world.

NovGauge is presented, a human-anchored benchmark for fine-grained novelty assessment diagnosis, and a cascading diagnostic pipeline that verifies per-dimension correctness, evidence grounding, and logical support is proposed.

Sci-MMR is introduced, a benchmark for multi-step evidence-grounded scientific reasoning built on structured argument graphs linking scientific claims, citation-grounded knowledge, visual evidence, and supporting regions, and it is found that current answer-centric benchmarks substantially overestimate the evidence-grounded reasoning capabilities of multimodal research agents.

A shift from task-specific models to a perception-centered architecture could revolutionize how AI agents provide ongoing, adaptive support in complex environments.

Low-probability tokens disproportionately influence model updates, and a simple reweighting strategy can significantly enhance performance without sacrificing generalization.

Self-improving search agents thrive when feedback and policy evolution are intertwined, leading to sustained performance gains and reduced hallucinations.

Natural-language critiques can transform how we evaluate and optimize song generation models, leading to more human-aligned outputs.

LLMs struggle with personal information retrieval, with the best model only achieving 57.3% accuracy on a new benchmark designed to evaluate mobile assistant capabilities.

IACM-RL reduces infinite loops and stale context errors by proactively managing dynamic user intents, setting a new standard for robust tool invocation.

H$^2$SD achieves superior reasoning performance by intelligently adapting teacher signals based on trajectory outcomes, leading to more effective learning in large language models.

Even state-of-the-art language models struggle significantly in real-world tasks, exposing critical shortcomings in their deployment readiness.

Even the top-performing language models struggle with archive-grounded reasoning, achieving only 59.4% accuracy on a benchmark designed to test their agentic capabilities across diverse workplace documents.

Today's best language models can barely make sense of your messy group chats and fragmented digital life, achieving only 19% accuracy on a new benchmark of real-world reasoning.

Coding agents can now evolve their own harnesses to outperform human-designed ones, thanks to a novel observability-driven approach.

Multi-turn reinforcement learning gets a boost: weighting trajectories by semantic similarity dramatically improves baseline estimation and agent performance in long-document visual QA.

Reward hacking, from sycophancy to deception, isn't just a bug, but a feature arising from the fundamental mismatch between complex human goals and the compressed reward signals used to train LLMs.

Even the best LLMs fail to follow complex constraints in tool use more than 50% of the time, revealing a critical weakness in real-world agent deployment.

RFT's impressive in-domain performance masks surprisingly weak generalization to new environments, highlighting a critical challenge for deploying LLM agents in the real world.

GPT-5's scientific reasoning skills plummet by nearly 50% when tackling multi-step workflows, revealing a critical gap in current LLM agents' ability to orchestrate complex tool use.

Retrofit your VLMs with Multi-Head Latent Attention (MLA) for faster inference and smaller memory footprint, without costly pretraining, using this parameter-efficient conversion framework.

Finally, a fully open-source, reproducible system for long-form song generation is here, complete with licensed data, code, and a Qwen-based model that rivals closed-source systems.