10 papers
COMET: Contrastive Motion-Enhanced Temporal Reasoning for Video Multimodal Large Language Models
Chenghua Zhu, Zhaolu Kang, Qifan Shi +8
Video multimodal large language models have advanced significantly, yet fine-grained motion-temporal understanding remains fragile. The core bottleneck is not only sparse frame sam…
Is This Your Final Answer? Cross-Contextual Consistency as a Measure of LLM Credibility
Siyang Wu, Yibo Jiang, Bryon Aragam
Large language models (LLMs) are powerful black-box systems, making it difficult to discern whether their answers reflect stable internal beliefs or superficial pattern matching. W…
Contemporary AI lacks the imagination to diverge or negate in science
Honglin Bao, Siyang Wu, Xiao Liu +3
Bold claims that AI will accelerate scientific discovery have raced ahead of evidence from working scientists, yet large-scale, scientist-in-the-loop evidence is scarce. Here we mo…
Narrative Flattening: How Post-Training Compresses Thematic, Affective, and Stylistic Variation in LLM Fiction
Zehan Li, Yutong Zhu, Siyang Wu +2
Large language models produce fluent fiction, yet their creative output is widely seen as flat. We ask where this quality originates in the training and whether it affects differen…
Zero-Forgetting CISS via Dual-Phase Cognitive Cascades
Yuquan Lu, Yifu Guo, Zishan Xu +6
Continual semantic segmentation (CSS) is a cornerstone task in computer vision that enables a large number of downstream applications, but faces the catastrophic forgetting challen…
EDIS: Diagnosing LLM Reasoning via Entropy Dynamics
Chenghua Zhu, Siyan Wu, Xiangkang Zeng +6
Entropy-based confidence signals are increasingly leveraged to improve reasoning in large language models (LLMs), yet existing approaches treat confidence as a static quantity -- t…
How do Language Models Generate Slang: A Systematic Comparison between Human and Machine-Generated Slang Usages
Siyang Wu, Zhewei Sun
Slang is a commonly used type of informal language that poses a daunting challenge to NLP systems. Recent advances in large language models (LLMs), however, have made the problem m…
Mapping Overlaps in Benchmarks through Perplexity in the Wild
Siyang Wu, Honglin Bao, Sida Li +2
We introduce benchmark signatures to characterize the capacity demands of LLM benchmarks and their overlaps. Signatures are sets of salient tokens from in-the-wild corpora whose mo…
Language Models Should be Used to Surface the Unwritten Code of Science and Society
Honglin Bao, Siyang Wu, Jiwoong Choi +2
This paper calls on the research community not only to investigate how human biases are inherited by large language models (LLMs) but also to explore how these biases in LLMs can b…
Missing vs. Unused Knowledge Hypothesis for Language Model Bottlenecks in Patent Understanding
Siyang Wu, Honglin Bao, Nadav Kunievsky +1
While large language models (LLMs) excel at factual recall, the real challenge lies in knowledge application. A gap persists between their ability to answer complex questions and t…