Papers by Victor Bursztyn
RaDA: Retrieval-augmented Web Agent Planning with LLMs (2024.findings-acl)
Copied to clipboard
| Challenge: | Agents powered by large language models inherit important limitations such as the restricted context length, dependency on human-engineered exemplars, and insufficient generalization. |
| Approach: | They propose a novel planning method for Web agents that disentangles planning into two stages: for a new given task, it decomposes tasks into high-level subtasks; and then iteratively synthesizes actions based on dynamically retrieved exemplars. |
| Outcome: | The proposed method decomposes tasks into high-level subtasks and iteratively synthesizes actions based on dynamically retrieved exemplars. |
Doc-React: Multi-page Heterogeneous Document Question-answering (2025.acl-short)
Copied to clipboard
Junda Wu, Yu Xia, Tong Yu, Xiang Chen, Sai Sree Harsha, Akash V Maharaj, Ruiyi Zhang, Victor Bursztyn, Sungchul Kim, Ryan A. Rossi, Julian McAuley, Yunyao Li, Ritwik Sinha
| Challenge: | Existing methods for integrating information across multiple modalities are suboptimal for multi-page, multimodal documents. |
| Approach: | They propose an adaptive iterative framework that balances information gain and uncertainty reduction at each step. |
| Outcome: | The proposed framework captures relevant multimodal content and achieves strong performance on complex QA tasks. |
Learning to Perform Complex Tasks through Compositional Fine-Tuning of Language Models (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Recent work on how to encode compositional task structure has been limited by semantic parsing and multihop reasoning for the purpose of Q&A. |
| Approach: | They propose an approach to decomposing a target task into component tasks and fine-tuning smaller LMs on a curriculum of such component tasks. |
| Outcome: | The proposed approach outperforms end-to-end learning even with equal data, and gets better as more component tasks are modeled. |
“It doesn’t look good for a date”: Transforming Critiques into Preferences for Conversational Recommendation Systems (2021.emnlp-main)
Copied to clipboard
| Challenge: | Conversational recommendation systems (CRSs) aim to refine options over multiple turns of a conversation, but they are not as flexible as real conversations. |
| Approach: | They propose a method for transforming a user critique into a positive preference . they use a large neural language model to perform critique-to-preference transformation . |
| Outcome: | The proposed method improves recommendations in restaurant domain using a new dataset of restaurant critiques. |
TeamFusion: Supporting Open-ended Teamwork with Multi-Agent Systems (2026.acl-long)
Copied to clipboard
| Challenge: | Many group decisions are open-ended, and aggregation approaches suppress minority perspectives . team members must surface hidden assumptions, discuss disagreements, negotiate acceptable trade-offs . |
| Approach: | They propose a multi-agent system that instantiates a proxy agent for each team member . they also conduct a structured discussion to elicit agreements and disagreements . |
| Outcome: | The proposed system outperforms direct aggregation on two teamwork tasks . it can judge how well individual views are represented in team decisions and consensually good deliverables . |
Disambiguation in Conversational Question Answering in the Era of LLMs and Agents: A Survey (2025.emnlp-main)
Copied to clipboard
Mehrab Tanjim, Yeonjun In, Xiang Chen, Victor Bursztyn, Ryan A. Rossi, Sungchul Kim, Guang-Jie Ren, Vaishnavi Muppala, Shun Jiang, Yongsung Kim, Chanyoung Park
| Challenge: | Existing literature on ambiguity and disambiguation with Large Language Models (LLMs) ambiguities are a fundamental challenge in human-AI interactions due to complexity and flexibility of human language. |
| Approach: | They propose to define key terms and concepts and categorize various disambiguation approaches enabled by LLMs and provide a comparative analysis of their advantages and disadvantages. |
| Outcome: | The proposed frameworks are compared against different disambiguation approaches and highlight their relevance for future research. |
A Flash in the Pan: Better Prompting Strategies to Deploy Out-of-the-Box LLMs as Conversational Recommendation Systems (2025.coling-main)
Copied to clipboard
Gustavo Adolpho Lucas de Carvalho, Simon Ben Igeri, Jennifer Healey, Victor Bursztyn, David Demeter, Lawrence A. Birnbaum
| Challenge: | Recent studies have shown that using conversation history can improve question generation and product recommendation in naturalistic, multi-round conversational recommendation settings. |
| Approach: | They propose a method to generate better questions to elicit human preferences and to make recommendations using the information gained through these questions. |
| Outcome: | The proposed method beats state-of-the-art benchmarks on two datasets and shows that it is more accurate when users answer more questions than prior methods. |
SQLSpace: A Representation Space for Text-to-SQL to Discover and Mitigate Robustness Gaps (2025.findings-emnlp)
Copied to clipboard
| Challenge: | SQLSpace is a representation for text-to-SQL examples derived with minimal human intervention. |
| Approach: | They introduce SQLSpace, a human-interpretable, generalizable, compact representation for text-to-SQL examples . they demonstrate that SQLSpace enables analysis that would be difficult with raw examples alone . |
| Outcome: | The proposed representations are human-interpretable and generalizable . they are used to evaluate models with a granularity beyond overall accuracy scores . |