5 papers
PACE: A Proxy for Agentic Capability Evaluation
Yueqi Song, Lintang Sutawika, Jiarui Liu +8
Evaluating LLM agents on benchmarks like SWE-Bench and GAIA can be expensive, time-consuming, and requires complex infrastructure. A single evaluation can cost thousands of dollars…
ReasonOps: Operator Segmentation for LLM Reasoning Traces
Daniel Lee, Owen Queen, James Zou
Chain-of-thought traces from large reasoning models can span tens of thousands of tokens, yet we lack a vocabulary for describing their internal structure. Previous methods develop…
Learning to Repair Lean Proofs from Compiler Feedback
Evan Wang, Simon Chess, Daniel Lee +4
As neural theorem provers become increasingly agentic, the ability to interpret and act on compiler feedback is critical. However, existing Lean datasets consist almost exclusively…
SweRank+: Multilingual, Multi-Turn Code Ranking for Software Issue Localization
Revanth Gangi Reddy, Ye Liu, Wenting Zhao +7
Maintaining large-scale, multilingual codebases hinges on accurately localizing issues, which requires mapping natural-language error descriptions to the relevant functions that ne…
WINELL: Wikipedia Never-Ending Updating with LLM Agents
Revanth Gangi Reddy, Tanay Dixit, Jiaxin Qin +7
Wikipedia, a vast and continuously consulted knowledge base, faces significant challenges in maintaining up-to-date content due to its reliance on manual human editors. Inspired by…