11 papers
WorldMemArena: Evaluating Multimodal Agent Memory Through Action-World Interaction
Chengzhi Liu, Yuzhe Yang, Sophia Xiao Pu +14
Multimodal large language models are increasingly deployed as long-horizon agents, where memory must do more than recall: it must track an evolving world, revise what has gone stal…
Deep FinResearch Bench: Evaluating AI's Ability to Conduct Professional Financial Investment Research
Mirazul Haque, Antony Papadimitriou, Samuel Mensah +6
We introduce Deep FinResearch Bench, a practical and comprehensive evaluation framework for deep research (DR) agents in financial investment research. The benchmark assesses three…
Detecting Non-Membership in LLM Training Data via Rank Correlations
Pranav Shetty, Mirazul Haque, Zhiqiang Ma +1
As large language models (LLMs) are trained on increasingly vast and opaque text corpora, determining which data contributed to training has become essential for copyright enforcem…
Distill and Align Decomposition for Enhanced Claim Verification
Jabez Magomere, Elena Kochkina, Samuel Mensah +6
Complex claim verification requires decomposing sentences into verifiable subclaims, yet existing methods struggle to align decomposition quality with verification performance. We…
ExStrucTiny: A Benchmark for Schema-Variable Structured Information Extraction from Document Images
Mathieu Sibue, Andres Muñoz Garza, Samuel Mensah +4
Enterprise documents, such as forms and reports, embed critical information for downstream applications like data archiving, automated workflows, and analytics. Although generalist…
Entropy-Gated Branching for Efficient Test-Time Reasoning
Xianzhi Li, Ethan Callanan, Abdellah Ghassel +1
Test-time compute methods can significantly improve the reasoning capabilities and problem-solving accuracy of large language models (LLMs). However, these approaches require subst…