1 citations · 3 across the 12 of their papers we have counts for
13 papers
KDFP: A first-principles approach to knowledge distillation in large language models
Ryan Swift, Konstantinos Psounis
Knowledge distillation is an established technique for improving the capabilities of small, efficient student models by training them with the representations of larger, more capab…
Agentic Context Learning with Self-Discovered Specification
Jike Zhong, Ming Li, Yuxiang Lai +8
Context learning is an emerging inference-time task where LLMs must learn and apply novel, task-specific knowledge from intricate contexts absent from pre-training; even frontier m…
From Shortcuts to Reasoning: Robust Post-Training of Theory of Mind with Reinforcement Learning
Jike Zhong, Yuxiang Lai, Ming Li +5
Theory of Mind (ToM) is a must-acquire skill for modern foundation model systems to operate effectively and safely in the real world. Recent works have explored honing ToM via post…
Agent-Sentry: Bounding LLM Agents via Execution Provenance
Rohan Sequeira, Stavros Damianakis, Umar Iqbal +1
Agentic computing systems, while immensely capable, raise serious security, privacy, and safety concerns. A key issue is that the full set of functionalities offered by these syste…
VRIQ: Benchmarking and Diagnosing the Visual-Reasoning IQ of Vision-Language Models
Tina Khezresmaeilzadeh, Jike Zhong, Konstantinos Psounis
Vision-language model (VLM) errors on visual reasoning may come from missing visual structure, from mistakes after that structure is extracted, or from both. We introduce VRIQ, a b…
TIR-Bench: A Comprehensive Benchmark for Agentic Thinking-with-Images Reasoning
Ming Li, Jike Zhong, Shitian Zhao +6
The frontier of visual reasoning is shifting toward models like OpenAI o3, which can intelligently create and operate tools to transform images for problem-solving, also known as t…
Preserving Privacy and Utility in LLM-Based Product Recommendations
Tina Khezresmaeilzadeh, Jiang Zhang, Dimitrios Andreadis +1
Large Language Model (LLM)-based recommendation systems leverage powerful language models to generate personalized suggestions by processing user interactions and preferences. Unli…
SpinML: Customized Synthetic Data Generation for Private Training of Specialized ML Models
Jiang Zhang, Rohan Xavier Sequeira, Konstantinos Psounis
Specialized machine learning (ML) models tailored to users needs and requests are increasingly being deployed on smart devices with cameras, to provide personalized intelligent ser…
CATS: A framework for Cooperative Autonomy Trust & Security
Namo Asavisanu, Tina Khezresmaeilzadeh, Rohan Sequeira +4
With cooperative perception, autonomous vehicles can wirelessly share sensor data and representations to overcome sensor occlusions, improving situational awareness. Securing such…
EEE-Bench: A Comprehensive Multimodal Electrical And Electronics Engineering Benchmark
Ming Li, Jike Zhong, Tianle Chen +2
Recent studies on large language models (LLMs) and large multimodal models (LMMs) have demonstrated promising skills in various domains including science and mathematics. However,…
Echoes of Privacy: Uncovering the Profiling Practices of Voice Assistants
Tina Khezresmaeilzadeh, Elaine Zhu, Kiersten Grieco +3
Many companies, including Google, Amazon, and Apple, offer voice assistants as a convenient solution for answering general voice queries and accessing their services. These voice a…
Differentially Private Federated Learning without Noise Addition: When is it Possible?
Jiang Zhang, Konstantinos Psounis, Salman Avestimehr
Federated Learning (FL) with Secure Aggregation (SA) has gained significant attention as a privacy preserving framework for training machine learning models while preventing the se…
Efficient Toxic Content Detection by Bootstrapping and Distilling Large Language Models
Jiang Zhang, Qiong Wu, Yiming Xu +3
Toxic content detection is crucial for online services to remove inappropriate content that violates community standards. To automate the detection process, prior works have propos…