19 papers
How Causality Bridges the Semantic Gap
Shuhao Zhang, Xuran Zhou, Han Guo +2
Numerical measurements capture how a system behaves, but often leave the meanings of its variables unspecified. Some variables are measured but never labeled, and others are never…
JudgeProfile: Understanding and Steering Subjectivity in LLM Judges
Qi Cao, Kangning Liu, Xuan Kan +10
LLM judges are inherently subjective, often favoring different responses in pairwise comparison when neither option is objectively wrong. To study this subjectivity, we introduce J…
OPBackdoor: Opportunistic Backdoors via Alibi-Aligned Reasoning
Eric Xue, Ruiyi Zhang, Kevin Xue +3
When a backdoor trigger activates the target response regardless of the triggered prompt context, the backdoor objective reveals itself. Challenging this trigger-sufficient formula…
The Missing Complement: State-Conditioned Minimal Sufficient Evidence for Coding Agents
Zhexi Feng, Ruiyi Zhang, Yongbo Yang +1
Retrieval assembles repository context by ranking passages for relevance to the current query. A coding agent halfway through an issue has already read much of what such a ranker r…
AIBuildAI-2.5: Efficient Autonomous AI Model Development Through LLM-Guided Tree Search
Peijia Qin, Ruiyi Zhang, Qi Cao +3
Autonomous agents that automatically build artificial intelligence (AI) models could broaden access to AI across science and engineering. A popular line of such agents frames model…
Can LLMs Take the Pulse of the Economy? A Real-Time Evaluation of LLM Nowcasts on Macroeconomic Indicators
Xinyue Zhao, Ruiyi Zhang, Liqin Ye +3
Nowcasting headline macroeconomic indicators, i.e., estimating an indicator's value for the current reference period before its official release, is critical for monetary policy an…
Reconstructing the Right Episode: Evaluating Interleaved Conversational Memory Beyond Long Context
Zhexi Feng, Ruiyi Zhang, Yongbo Yang +1
Conversations with chat assistants increasingly span many topics in a single long-running thread, challenging memory systems. Existing long-context and memory benchmarks often expo…
ATLAS: Agentic Test-time Learning-to-Allocate Scaling
Peijia Qin, Qi Cao, Pengtao Xie
Test-time scaling has become a major way to improve large language model reasoning, but its orchestration has remained designer-engineered: a fixed sample budget, a fixed refinemen…
AIBuildAI-2: A Knowledge-Enhanced Agent for Automatically Building AI Models
Ruiyi Zhang, Peijia Qin, Qi Cao +2
AI models underpin data-centric applications from image and text processing to scientific discovery in biology, physics, and chemistry. Yet developing them remains heavily manual,…
AutoResearch AI: Towards AI-Powered Research Automation for Scientific Discovery
Guiyao Tie, Jiawen Shi, Dingjie Song +20
Scientific research is being reshaped by AI systems that move beyond isolated assistance toward longer-horizon workflows spanning literature grounding, hypothesis generation, exper…
LLMs Know When They Know, but Do Not Act on It: A Metacognitive Harness for Test-time Scaling
Qi Cao, Yufan Wang, Peijia Qin +2
Large language models (LLMs) often expose useful signals of self-monitoring: before solving a problem, they can estimate whether they are likely to succeed, and after solving it, t…
BioTool: A Comprehensive Tool-Calling Dataset for Enhancing Biomedical Capabilities of Large Language Models
Xin Gao, Ruiyi Zhang, Meixi Du +2
Despite the success of large language models (LLMs) on general-purpose tasks, their performance in highly specialized domains such as biomedicine remains unsatisfactory. A key limi…
Send a SCOUT First: Pre-hoc Reasoning for Adaptive Detector Allocation in Prompt-Injection Defense
Shuhao Zhang, Jiarui Li, Qi Cao +2
Prompt-injection detectors are heterogeneous: each is strong on a different slice of attacks, and none is always reliable. Yet existing systems still treat detection as a fixed sin…
AIBuildAI: An AI Agent for Automatically Building AI Models
Ruiyi Zhang, Peijia Qin, Qi Cao +2
AI models underpin modern intelligent systems, driving advances across science, medicine, finance, and technology. Yet developing high-performing AI models remains a labor-intensiv…
Models Under SCOPE: Scalable and Controllable Routing via Pre-hoc Reasoning
Qi Cao, Shuhao Zhang, Ruizhe Zhou +3
Model routing chooses which language model to use for each query. By sending easy queries to cheaper models and hard queries to stronger ones, it can significantly reduce inference…
FunPRM: Function-as-Step Process Reward Model with Meta Reward Correction for Code Generation
Ruiyi Zhang, Peijia Qin, Qi Cao +2
Code generation is a core application of large language models (LLMs), yet LLMs still frequently fail on complex programming tasks. Given its success in mathematical reasoning, tes…
JURIST: Training Small Data-Reweighted Reasoning Judges for Test-Time Scaling in Code Generation
Peijia Qin, Ruiyi Zhang, Qi Cao +1
Best-of-N selection for code generation is only as good as the judge that picks among the candidates, and the judges accurate enough for this role are costly frontier LLMs. A small…
DreamPRM-Code: Function-as-Step Process Reward Model with Label Correction for LLM Coding
Ruiyi Zhang, Peijia Qin, Qi Cao +1
Process Reward Models (PRMs) have become essential for improving Large Language Models (LLMs) via test-time scaling, yet their effectiveness in coding remains limited due to the la…
DreamPRM-1.5: Unlocking the Potential of Each Instance for Multimodal Process Reward Model Training
Qi Cao, Pengtao Xie
Training multimodal process reward models (PRMs) is hard due to (i) distribution shift between training set and test set and (ii) quality imbalance across training data samples. Wh…