Tag
The paper introduces Org-Agent, a unified constraint-centric reasoning framework that enables LLM agents to coordinate requests from multiple users and use knowledge distributed across interactions, decomposing tasks via dependency graphs and topological scheduling. Experiments on MUSES-Bench and GroupMemBench show gains in cross-user decision-making and memory capabilities.
MolDesignBench is a new benchmark for evaluating LLM-based agents in scenario-grounded molecular design, comprising 2K instances with implicit and explicit constraints, revealing that current frontier LLMs achieve low success rates, especially in reasoning about implicit constraints and infeasibility detection.
This paper proposes a neuro-symbolic approach that integrates a MaxSAT oracle as a consistency validator to guide Vision-Language Models (VLMs) in solving Sudoku puzzles, improving logical consistency and the number of solved instances.
This paper introduces satisfiable drift, a failure mode where multi-turn reasoning systems silently violate prior commitments while maintaining internal logical consistency, dominating contradictions. The authors present DRIFT-Bench, a benchmark of 816 problems, and find that after repair, 98-100% of residual errors are drift errors.