15 papers
A mechanistic study of language model introspection
Jiahong Zou, Xiangkun Sun, Lingkai Kong +1
Large language models (LLMs) can sometimes report perturbations to their internal activations---even when the input provides no evidence that an intervention occurred. How do model…
Loyal Agents: Training LLM Agents to Protect Principal Interests Under Strategic Information Asymmetry
Zimeng Huang, Shilei Chen, Jiatong Zhao +2
As LLMs increasingly act as delegated agents, they are expected to protect principals' interests when interacting with external parties. Standard alignment objectives, such as help…
Every Cache Entry Earns Its Place: Global Allocation of Resolution and Coverage for KV Cache Compression
Haolin Tian, Yuzhe Liu, Tonghan Wang
As large language models (LLMs) process increasingly long contexts, KV cache storage and repeated access have become a major bottleneck. Existing KV cache compression methods rely…
Evaluating and Pricing Advertisements in AI-Generated Responses
John L. Turner-Smith, Zimeng Huang, Yuhan Fu +2
As search increasingly shifts toward LLM-driven answer engines, advertising is becoming embedded within the generated response itself and should therefore be evaluated for both use…
PILA: Plug-and-Play Insertion for LLM-native Advertising
Zhaowei Zhang, Yuhan Fu, Yihang Zhang +6
How to monetize large language models (LLMs) by naturally integrating sponsored content into their responses, known as LLM-native advertising, has recently emerged as a critical pr…
Duality for Optimal Multi-Item, Multi-Bidder Auction Design: Revenue Certificates through Deep Learning
Yanchen Jiang, David C. Parkes, Tonghan Wang
Characterizing revenue-optimal auctions for multi-item, multi-bidder settings remains a fundamental open problem, with no known closed-form solution existing beyond restrictive bin…
NaiAD: Initiate Data-Driven Research for LLM Advertising
Yihang Zhang, Zimeng Huang, Ren Zhai +2
Reconciling platform revenue with user experience in LLM advertising motivates a data-centric foundation. We introduce NaiAD, the first comprehensive dataset for LLM-native adverti…
How LLMs Are Persuaded: A Few Attention Heads, Rerouted
Xiangkun Sun, Lingkai Kong, Aoqi Zhang +2
Large language models can answer a factual question correctly, yet switch to an incorrect answer when exposed to persuasive context. Why this happens inside the model remains uncle…
LLM Advertisement based on Neuron Auctions
Peiran Yun, Wenxin Xu, Jiayuan Liu +4
As Large Language Models (LLMs) transition into conversational agents, generative advertising emerges as a crucial monetization strategy. However, embedding advertisements within u…
Incentive-Aware Multi-Fidelity Optimization for Generative Advertising in Large Language Models
Jiayuan Liu, Barry Wang, Jiarui Gan +4
Generative advertising in large language model (LLM) responses requires optimizing sponsorship configurations under two strict constraints: the strategic behavior of advertisers an…
LLM Active Alignment: A Nash Equilibrium Perspective
Tonghan Wang, Yuqi Pan, Xinyi Yang +3
We develop a game-theoretic framework for predicting and steering the behavior of populations of large language models (LLMs) through Nash equilibrium (NE) analysis. To avoid the i…
Composite Flow Matching for Reinforcement Learning with Shifted-Dynamics Data
Lingkai Kong, Haichuan Wang, Tonghan Wang +2
Incorporating pre-collected offline data can substantially improve the sample efficiency of reinforcement learning (RL), but its benefits can break down when the transition dynamic…
Adaptive Frontier Exploration on Graphs with Applications to Network-Based Disease Testing
Davin Choo, Yuqi Pan, Tonghan Wang +3
We study a sequential decision-making problem on a -node graph where each node has an unknown label from a finite set , drawn from a joint distribution $…
Robust Optimization with Diffusion Models for Green Security
Lingkai Kong, Haichuan Wang, Yuqi Pan +6
In green security, defenders must forecast adversarial behavior, such as poaching, illegal logging, and illegal fishing, to plan effective patrols. These behavior are often highly…
Translate Policy to Language: Flow Matching Generated Rewards for LLM Explanations
Xinyi Yang, Liang Zeng, Heng Dong +6
As humans increasingly share environments with diverse agents powered by RL, LLMs, and beyond, the ability to explain agent policies in natural language is vital for reliable coexi…