25 papers
LiveBrowseComp: Are Search Agents Searching, or Just Verifying What They Already Know?
HuiMing Fan, Xiao Wang, Zheng Chu +5
Are LLM-based search agents genuinely searching, or using the web to verify what they already know? We study this question on BrowseComp with three diagnostics. Our analysis reveal…
Parameter- and Bandwidth-Efficient Edge--cloud Many-to-Many Speech-to-Text Translation
Yexing Du, Kaiyuan Liu, Youcheng Pan +5
Multimodal large language models (MLLMs) have demonstrated significant potential for speech-to-text translation (S2TT). However, existing deployment paradigms face critical challen…
The Readout Shortcut: Positional Number Copying Dominates Arithmetic CoT Readout in Small Language Models
Ming Liu
Chain-of-thought (CoT) prompting is necessary for arithmetic in small language models, yet shuffling its steps preserves most performance. What does CoT contribute if not logical s…
Self-Training Doesn't Flatten Language -- It Restructures It: Surface Markers Amplify While Deep Syntax Dies
Ming Liu
Successive self-training on a language model's own outputs is widely characterized as a process of flattening: diversity drops, distributions narrow, and the text becomes "more lik…
EGL-SCA: Structural Credit Assignment for Co-Evolving Instructions and Tools in Graph Reasoning Agents
Zike Yuan, Yukun Cao, Han Zhang +7
Graph reasoning agents operating from natural-language inputs must solve a coupled problem: they must reconstruct a structured graph instance from text, decide whether existing com…
Perception Without Engagement: Dissecting the Causal Discovery Deficit in LMMs
Jiafeng Liang, Zhihao Zhu, Zihan Zhang +7
Although Large Multimodal Models (LMMs) have achieved strong performance on general video understanding, their susceptibility to textual prior shortcuts during causal discovery has…