3 papers
cs.CL2026
Before the Rollout Ends: Early Terminal Reward Prediction for Long-horizon Coding Agents
Jihan Yao, Sihan Zeng, Shangbin Feng +3
Long-horizon coding agents receive verifiable rewards only after completing expensive sequences of tool calls. This increases inference cost, amplifies early wrong hypotheses, and…
cs.SE2025
BigCodeArena: Unveiling More Reliable Human Preferences in Code Generation via Execution
Terry Yue Zhuo, Xiaolong Jin, Hange Liu +37
Crowdsourced model evaluation platforms, such as Chatbot Arena, enable real-time evaluation from human perspectives to assess the quality of model responses. In the coding domain,…
cs.AI2025
MMMG: a Comprehensive and Reliable Benchmark for Multitask Multimodal Generation
Jihan Yao, Yushi Hu, Yujie Yi +10
Automatically evaluating multimodal generation presents a significant challenge, as automated metrics often struggle to align with human evaluation, especially for complex tasks th…