most citedWhy Do Open-Source LLMs Struggle with Data Analysis? A Systematic Empirical Study

1 citations · 1 across the 9 of their papers we have counts for

collaborators
Showing cs.CLShow all

9 papers · 1 filter

cs.CL2026

Low Perplexity is Repetition: A One-Dimensional Self-Conditioning Attractor in Continuous Diffusion LMs

Shuai Zhang, Zijie Chen, Hongliang He +2

Continuous diffusion language models such as ELF report record-low generative perplexity (Gen-PPL). We find a catch: these models repeat far more than human text, and Gen-PPL rewar…

cs.CL2026

Rewarding the Scientific Process: Process-Level Reward Modeling for Agentic Data Analysis

Zhisong Qiu, Shuofei Qiao, Kewei Xu +4

Process Reward Models (PRMs) have achieved remarkable success in augmenting the reasoning capabilities of Large Language Models (LLMs) within static domains such as mathematics. Ho…

cs.CL2026

StructMem: Structured Memory for Long-Horizon Behavior in LLMs

Buqiang Xu, Yijun Chen, Jizhan Fang +5

Long-term conversational agents need memory systems that capture relationships between events, not merely isolated facts, to support temporal reasoning and multi-hop question answe…

cs.CL2026

What Makes AI Research Replicable? Executable Knowledge Graphs as Scientific Knowledge Representations

Yujie Luo, Zhuoyun Yu, Xuehai Wang +6

Replicating AI research is a crucial yet challenging task for large language model (LLM) agents. Existing approaches often struggle to generate executable code, primarily due to in…

cs.CL2026

Can We Predict Before Executing Machine Learning Agents?

Jingsheng Zheng, Jintian Zhang, Yujie Luo +5

Autonomous machine learning agents have revolutionized scientific discovery, yet they remain constrained by a Generate-Execute-Feedback paradigm. Previous approaches suffer from a…

cs.CL2026

InnoGym: Benchmarking the Innovation Potential of AI Agents

Jintian Zhang, Kewei Xu, Jingsheng Zheng +10

LLMs and Agents have achieved impressive progress in code generation, mathematical reasoning, and scientific discovery. However, existing benchmarks primarily measure correctness,…