10 papers
The Low Frequency Trap: Video Language Models Fail at Simple Event Bookkeeping
Sarvesh Baskar, Zikui Cai, Shayan Shabihi +5
Real-world video benchmarks provide broad coverage, but their fixed clips entangle event count, rate, duration, and visual complexity, making failure modes hard to isolate. While e…
Beyond Episodic Evaluation: Memory Architectural Bottlenecks in Sequential Embodied Question Answering
Zikui Cai, Kaushal Janga, Tan Dat Dao +15
Embodied question answering (EQA) is traditionally evaluated under an episodic formulation, where agents solve each task independently and reset internal state between episodes. Ho…
Compositional Adversarial Training for Robust Visual Watermarking
Anirudh Satheesh, Michael-Andrei Panaitescu-Liess, Andrew Xu +4
Robust watermarking is typically trained with random post-processing augmentation, but random sampling under-covers the combinatorial space of realistic attack pipelines and rarely…
TraceGen: World Modeling in 3D Trace Space Enables Learning from Cross-Embodiment Videos
Seungjae Lee, Yoonkyo Jung, Inkook Chun +8
Learning new robot tasks on new platforms and in new scenes from only a handful of demonstrations remains challenging. While videos of other embodiments - humans and different robo…
Zebra-CoT: A Dataset for Interleaved Vision Language Reasoning
Ang Li, Charles Wang, Deqing Fu +9
Humans often use visual aids, for example diagrams or sketches, when solving complex problems. Training multimodal models to do the same, known as Visual Chain of Thought (Visual C…
MORSE-500: A Programmatically Controllable Video Benchmark to Stress-Test Multimodal Reasoning
Zikui Cai, Andrew Wang, Anirudh Satheesh +10
Despite rapid advances in vision-language models (VLMs), current benchmarks for multimodal reasoning fall short in three key dimensions. First, they overwhelmingly rely on static i…
Jailbreak Distillation: Renewable Safety Benchmarking
Jingyu Zhang, Ahmed Elgohary, Xiawei Wang +5
Large language models (LLMs) are rapidly deployed in critical applications, raising urgent needs for robust safety benchmarking. We propose Jailbreak Distillation (JBDistill), a no…
Zero-Shot Vision Encoder Grafting via LLM Surrogates
Kaiyu Yue, Vasu Singla, Menglin Jia +6
Vision language models (VLMs) typically pair a modestly sized vision encoder with a large language model (LLM), e.g., Llama-70B, making the decoder the primary computational burden…
AegisLLM: Scaling Agentic Systems for Self-Reflective Defense in LLM Security
Zikui Cai, Shayan Shabihi, Bang An +5
We introduce AegisLLM, a cooperative multi-agent defense against adversarial attacks and information leakage. In AegisLLM, a structured workflow of autonomous agents - orchestrator…
Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities
Zora Che, Stephen Casper, Robert Kirk +12
Evaluations of large language model (LLM) risks and capabilities are increasingly being incorporated into AI risk management and governance frameworks. Currently, most risk evaluat…