Category
15,965 papers
This work introduces SAKI (Supervision Allocation with KL-constrained Interpolation), which combines a KL-constrained teacher-guided rollout with maximal coupling and reuses realized accept/correction events to route token-level supervision.
Pretrained transformers use little of their depth to follow references in context. Thirteen base models reliably follow only 1.4-3.6 lines, and extra pretrained loops add little. A task-trained rank-8 LoRA at one early layer extends this computation with all model weights frozen. Qwen3-8B improves from 15.5% to 99% exa...
Query-agnostic streaming video understanding requires vision-language models to continuously compress an indefinitely growing visual stream into a bounded memory before future queries are known. The performance depends critically on the memory mechanism--what observations to preserve, how to represent and consolidate t...
SafeCoEvo is proposed, a test-time Harness-Guard co-evolution framework for LLM agent safety that enables the external safety system to continually adapt from accumulated runtime experience to achieve simultaneous gains in safety and task utility.
Reach audiences
Advertise in front of researchers, engineers, and readers.
Agentic systems are now being widely used to orchestrate tools and reason over long contexts. However, the improving capabilities of the large language models powering these agents also create new attack surfaces for indirect prompt injection. In particular, an attacker may not need to place a complete malicious instru...
Persistence-Aware Credit Gating (PACG), which attenuates positive credit for unusually persistent visual claims and leaves all other credit unchanged, which extends to a larger model, a newer backbone, and the accuracy of HallusionBench also improves consistently.
MAADBench is presented, the first refreshable MAS AD benchmark designed for diverse, evolving LLM backbones underlying the agents, and automated provision of cost-free, deterministic step-level labels for fine-grained AD evaluation.
Agentic reinforcement learning (ARL) with verifiable rewards improves the ability of large language models (LLMs) to tackle knowledge-intensive tasks by learning to interleave search and reasoning. However, most existing ARL methods optimize only LLM-generated tokens and treat retrieved evidence as environment observat...
A pretrained text-to audiovisual generation model is adapted through source-conditioned feature modulation to jointly learn object addition and removal, and quantitative and qualitative evaluations demonstrate effective audiovisual object removal and addition.
Two key design dimensions of human-AI collaboration (autonomy and initiative) are examined and the collaboration patterns that they generate are explored, including Instruction, Delegation, Assistance, and Co-creation.
A nonstationary Lagrangian framework that constructs a policy sequence from solutions of unconstrained games or cooperative control problems, and proves a reward-independent feasibility certificate, cooperative feasibility and approximate optimality against feasible policy mixtures, and an extension to unbiased sampled...
This work introduces an algorithm that identifies a small, relevant subset of features for each subgoal, yielding options that generalize broadly and accelerate exploration in three sparse-reward, image-based domains, including the Atari game MontezumasRevenge.
Machine-learning networks can learn without detailed instructions from a designer. Those that do may offer insight into how the brain reorganizes itself during learning.
Announcing Gemini 4 Argon, our frontier model for real-world coding, enterprise knowledge work and cyber defense, rolling out soon.
With $2.1 million funding from Google.org, the open-source Public Transit Intelligence Hub will unify public transit monitoring, operations, and passenger communication.
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
Professor Sherry Turkle’s new book, “Artificial Intimacy,” offers a withering critique of chatbots and the antisocial dynamics she believes they encourage.
Machine Intelligence
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
