6 citations · 7 across the 11 of their papers we have counts for
7 papers · 1 filter
Divide and Inject: Can Agents Reconstruct an Indirect Prompt Injection from Fragments?
Michael Lee, Zhipeng Wei, Yue Dong +1
Agentic systems are now being widely used to orchestrate tools and reason over long contexts. However, the improving capabilities of the large language models powering these agents…
SCP-NL2TL: Selective Conformal Prediction with Semantic Verification for Natural Language to Temporal Logic Specifications
Yixuan Wang, Licheng Luo, Yu Fu +3
Translating natural language instructions into machine-interpretable formal specifications enables robots and autonomous systems to plan, reason, and formally verify their behavior…
SCOReD: Student-Aware CoT Optimization for Recommendation Distillation
Haz Sameen Shahgir, Yufei Li, Xiaohan Wei +8
Chain-of-thought (CoT) distillation in the recommendation domain is a necessary precursor to RL training, but raw teacher traces are ill-suited to this task. Large teachers approac…
Multi-modal Rail Crossing Safety Analysis
Paimon Goulart, Chansong Lim, Nícolas Roque dos Santos +4
Given one or more images of a railway crossing, can we leverage visual cues that allow us to robustly estimate how safe it is? Can we improve our ability to do so by introducing st…
D-Judge: Disrupting Multi-Turn Jailbreaks using Semantics-Preserving Output Rewriting
Huanli Gong, Zhipeng Wei, Yu Fu +4
Multi-turn jailbreak attacks pose a growing threat to large language model (LLM) safety because they exploit feedback from auxiliary judge models to iteratively refine prompts towa…
VLMs Need Words: Vision Language Models Ignore Visual Detail In Favor of Semantic Anchors
Haz Sameen Shahgir, Xiaofu Chen, Yu Fu +4
Vision-language models (VLMs) have achieved impressive performance across a wide range of multimodal tasks. However, they often fail on tasks that require fine-grained visual perce…
Do Reasoning LLMs Refuse What They Infer in Long Contexts?
Yu Fu, Haz Sameen Shahgir, Huanli Gong +3
Long-context LLMs can infer objectives that are not stated explicitly. This capability is useful for reasoning over documents, code, retrieved evidence, and tool traces, but it als…