The paper examines how tool-using agents fail in the evidence-to-action chain, identifying that failures often occur before execution due to incomplete investigation or acting without required evidence. Even after evidence is obtained, multi-action workflows face issues with unresolved prerequisites and incomplete execution. The study introduces SafeActBench, a benchmark with 656 cases across six domains, to analyze these failures.
The chart will appear after repeat observations. The current metric comes from the source.
3 source votesReal observations only. History before source connection is not reconstructed.
This research helps understand why tool-using agents fail in real-world tasks, highlighting issues in evidence collection and action execution that could inform better design of autonomous systems.
RealCompanion introduces a benchmark for evaluating AI's ability to understand humans through long-term conversations, featuring 27,218 messages from 10 real relationships. The study reveals that most queries don't require long-term memory and that current methods struggle to identify when memory is needed.
6 days agoDaily PapersThe paper investigates on-policy distillation (OPD) in cross-tokenizer settings, focusing on alignment coverage and supervision reliability. It finds that strict 1:1 token alignment covers most student-generated tokens despite vocabulary mismatches, and that restricting reverse KL to a top-16 subset of shared vocabulary achieves comparable accuracy to full shared-vocabulary OPD, while adding span supervision reduces accuracy.
yesterdayDaily PapersThe paper introduces WING, a framework that transfers interaction knowledge from human egocentric videos to robot policies by separating observer-induced motion from hand-object interactions and using spectral analysis to identify shared temporal structures between human and robot behaviors. It achieves high success rates on multiple benchmarks and demonstrates strong performance in real-world tasks.
5 days agoDaily PapersThis paper introduces ProWAM, a progressive world action model that predicts actions and sparse visual sub-goals to guide robotic control. It improves long-horizon planning by using a single video-backbone pass for efficient action generation and demonstrates strong performance on simulation and real-world benchmarks.
6 days agoDaily PapersEgo2Act is a benchmark for evaluating goal-directed manipulation in egocentric video generation, featuring 2,640 videos from 110 real-world tasks. It assesses whether video generation models can create realistic egocentric videos of a hand performing multi-step object manipulations to achieve high-level goals.
6 days agoLessWrongThe text discusses a scenario where an AI must respond to the question 'What's the date?' without access to real-time data. It explores the challenge of providing an accurate date without hallucinating, considering OpenAI's training focus on avoiding false information. The text speculates on potential model versions and knowledge cutoff dates but acknowledges uncertainty due to lack of explicit information.
6 days ago