154 citations · 1k across the 49 of their papers we have counts for
53 papers
Synthetic Dialogue Generation for Interactive Conversational Elicitation & Recommendation (ICER)
Moonkyung Ryu, Chih-Wei Hsu, Yinlam Chow +2
While language models (LMs) offer great potential for conversational recommender systems (CRSs), the paucity of public CRS data makes fine-tuning LMs for CRSs challenging. In respo…
Descriptive History Representations: Learning Representations by Answering Questions
Guy Tennenholtz, Jihwan Jeong, Chih-Wei Hsu +2
Effective decision making in partially observable environments requires compressing long interaction histories into informative representations. We introduce Descriptive History Re…
Gemma 3 Technical Report
Gemma Team, Aishwarya Kamath, Johan Ferret +209
We introduce Gemma 3, a multimodal addition to the Gemma family of lightweight open models, ranging in scale from 1 to 27 billion parameters. This version introduces vision underst…
Inference-Aware Fine-Tuning for Best-of-N Sampling in Large Language Models
Yinlam Chow, Guy Tennenholtz, Izzeddin Gur +7
Recent studies have indicated that effectively utilizing inference-time compute is crucial for attaining better performance from large language models (LLMs). In this work, we prop…
Preference Adaptive and Sequential Text-to-Image Generation
Ofir Nabati, Guy Tennenholtz, ChihWei Hsu +5
We address the problem of interactive text-to-image (T2I) generation, designing a reinforcement learning (RL) agent which iteratively improves a set of generated images for a user…
Embedding-Aligned Language Models
Guy Tennenholtz, Yinlam Chow, Chih-Wei Hsu +3
We propose a novel approach for training large language models (LLMs) to adhere to objectives defined within a latent embedding space. Our method leverages reinforcement learning (…
DynaMITE-RL: A Dynamic Model for Improved Temporal Meta-Reinforcement Learning
Anthony Liang, Guy Tennenholtz, Chih-wei Hsu +3
We introduce DynaMITE-RL, a meta-reinforcement learning (meta-RL) approach to approximate inference in environments where the latent state evolves at varying rates. We model episod…
Preference Elicitation with Soft Attributes in Interactive Recommendation
Erdem Biyik, Fan Yao, Yinlam Chow +4
Preference elicitation plays a central role in interactive recommender systems. Most preference elicitation approaches use either item queries that ask users to select preferred it…
Factual and Personalized Recommendations using Language Models and Reinforcement Learning
Jihwan Jeong, Yinlam Chow, Guy Tennenholtz +4
Recommender systems (RSs) play a central role in connecting users to content, products, and services, matching candidate items to users based on their preferences. While traditiona…
Demystifying Embedding Spaces using Large Language Models
Guy Tennenholtz, Yinlam Chow, Chih-Wei Hsu +6
Embeddings have become a pivotal means to represent complex, multi-faceted information about entities, concepts, and relationships in a condensed and useful format. Nevertheless, t…
Offline Reinforcement Learning for Mixture-of-Expert Dialogue Management
Dhawal Gupta, Yinlam Chow, Aza Tulepbergenov +2
Reinforcement learning (RL) has shown great promise for developing dialogue management (DM) agents that are non-myopic, conduct rich conversations, and maximize overall user satisf…
Dynamic Planning in Open-Ended Dialogue using Reinforcement Learning
Deborah Cohen, Moonkyung Ryu, Yinlam Chow +8
Despite recent advances in natural language understanding and generation, and decades of research on the development of conversational bots, building automated agents that can carr…
A Mixture-of-Expert Approach to RL-based Dialogue Management
Yinlam Chow, Aza Tulepbergenov, Ofir Nachum +3
Despite recent advancements in language models (LMs), their application to dialogue management (DM) problems and ability to carry on rich conversations remain a challenge. We use r…
Efficient Risk-Averse Reinforcement Learning
Ido Greenberg, Yinlam Chow, Mohammad Ghavamzadeh +1
In risk-averse reinforcement learning (RL), the goal is to optimize some risk measure of the returns. A risk measure often focuses on the worst returns out of the agent's experienc…
SAFER: Data-Efficient and Safe Reinforcement Learning via Skill Acquisition
Dylan Slack, Yinlam Chow, Bo Dai +1
Methods that extract policy primitives from offline demonstrations using deep generative models have shown promise at accelerating reinforcement learning(RL) for new tasks. Intuiti…
Non-Stationary Latent Bandits
Joey Hong, Branislav Kveton, Manzil Zaheer +4
Users of recommender systems often behave in a non-stationary fashion, due to their evolving preferences and tastes over time. In this work, we propose a practical approach for fas…
CoinDICE: Off-Policy Confidence Interval Estimation
Bo Dai, Ofir Nachum, Yinlam Chow +3
We study high-confidence behavior-agnostic off-policy evaluation in reinforcement learning, where the goal is to estimate a confidence interval on a target policy's value, given on…
Safe Reinforcement Learning with Natural Language Constraints
Tsung-Yen Yang, Michael Hu, Yinlam Chow +2
While safe reinforcement learning (RL) holds great promise for many practical applications like robotics or autonomous cars, current approaches require specifying constraints in ma…
Control-Aware Representations for Model-based Reinforcement Learning
Brandon Cui, Yinlam Chow, Mohammad Ghavamzadeh
A major challenge in modern reinforcement learning (RL) is efficient control of dynamical systems from high-dimensional sensory observations. Learning controllable embedding (LCE)…
Variational Model-based Policy Optimization
Yinlam Chow, Brandon Cui, MoonKyung Ryu +1
Model-based reinforcement learning (RL) algorithms allow us to combine model-generated data with those collected from interaction with the real system in order to alleviate the dat…
Latent Bandits Revisited
Joey Hong, Branislav Kveton, Manzil Zaheer +3
A latent bandit problem is one in which the learning agent knows the arm reward distributions conditioned on an unknown discrete latent state. The primary goal of the agent is to i…
Non-Stationary Off-Policy Optimization
Joey Hong, Branislav Kveton, Manzil Zaheer +2
Off-policy learning is a framework for evaluating and optimizing policies without deploying them, from data collected by another policy. Real-world environments are typically non-s…
Predictive Coding for Locally-Linear Control
Rui Shu, Tung Nguyen, Yinlam Chow +5
High-dimensional observations and unknown dynamics are major challenges when applying optimal control to many real-world decision making tasks. The Learning Controllable Embedding…
AlgaeDICE: Policy Gradient from Arbitrary Experience
Ofir Nachum, Bo Dai, Ilya Kostrikov +3
In many real-world applications of reinforcement learning (RL), interactions with the environment are limited due to cost or feasibility. This presents a challenge to traditional R…