Papers by Hamed Zamani
Personalized Generation In Large Model Era: A Survey (2025.acl-long)
Copied to clipboard
Yiyan Xu, Jinghao Zhang, Alireza Salemi, Xinting Hu, Wenjie Wang, Fuli Feng, Hamed Zamani, Xiangnan He, Tat-Seng Chua
| Challenge: | Recent advances in large generative models have catalyzed a paradigm shift in content generation to Personalized Generation (PGen). |
| Approach: | They propose a multi-level taxonomy that systematically formalizes PGen's key components, core objectives, and abstract workflows. |
| Outcome: | The proposed taxonomy bridging PGen research across multiple modalities highlights open challenges and promising directions for future exploration. |
LaMP-QA: A Benchmark for Personalized Long-form Question Answering (2025.emnlp-main)
Copied to clipboard
| Challenge: | Personalization in question answering systems remains underexplored due to lack of resources . a new benchmark for personalized answer generation is being developed . |
| Approach: | They propose a LaMP-QA benchmark for evaluating personalized answer generation . they compare evaluation strategies for generating personalized responses and measure their alignment with human preferences. |
| Outcome: | The proposed benchmark shows that incorporating the personalized context provided leads to performance improvements. |
DISAPERE: A Dataset for Discourse Structure in Peer Review Discussions (2022.naacl-main)
Copied to clipboard
Neha Kennard, Tim O’Gorman, Rajarshi Das, Akshay Sharma, Chhandak Bagchi, Matthew Clinton, Pranay Kumar Yelugam, Hamed Zamani, Andrew McCallum
| Challenge: | Prior work on labeling arguments extracted from peer review text has focused qualified labor force on labelling arguments extracted by the text. |
| Approach: | They synthesize label sets from prior work and extend them to include fine-grained annotations of review and rebuttal sentences. |
| Outcome: | The proposed dataset synthesizes label sets from prior work and extends them to include fine-grained annotation of review and rebuttal sentences. |
ICXML: An In-Context Learning Framework for Zero-Shot Extreme Multi-Label Classification (2024.findings-naacl)
Copied to clipboard
| Challenge: | Existing research has focused on fully supervised XMC, but real-world scenarios often lack supervision signals, highlighting the importance of zero-shot settings. |
| Approach: | They propose a framework that generates a set of candidate labels through in-context learning and then reranks them. |
| Outcome: | The proposed framework advances state-of-the-art on two diverse public benchmarks. |
Multi-step Entity-centric Information Retrieval for Multi-Hop Question Answering (D19-58)
Copied to clipboard
Rajarshi Das, Ameya Godbole, Dilip Kavarthapu, Zhiyu Gong, Abhishek Singhal, Mo Yu, Xiaoxiao Guo, Tian Gao, Hamed Zamani, Manzil Zaheer, Andrew McCallum
| Challenge: | Multi-hop question answering (QA) requires an information retrieval system that can find multiple supporting evidence needed to answer the question. |
| Approach: | They propose a technique that uses information of entities present in the initial retrieved evidence to learn to ‘hop’ onto other relevant evidence. |
| Outcome: | The proposed method boosts retrieval performance on a multi-hop question answering dataset with 5 million Wikipedia paragraphs and a model without training increases its performance by 10.59 F1. |
Predicting Prerequisite Relations for Unseen Concepts (2022.emnlp-main)
Copied to clipboard
| Challenge: | Concept prerequisite learning (CPL) is a task of building a concept graph by structuring open knowledge in prerequisite relations. |
| Approach: | They propose to use both content-based and graph-based models to build a concept graph by structuring open knowledge in prerequisite relations. |
| Outcome: | The proposed approach improves F1 scores by 10% on three public benchmarks. |
LaMP: When Large Language Models Meet Personalization (2024.acl-long)
Copied to clipboard
| Challenge: | Existing benchmarks for personalization in large language models are understudied . |
| Approach: | They propose a benchmark for training and evaluating language models for producing personalized outputs using a set of seven personalized tasks . they propose two retrieval augmentation approaches that retrieve personal items from each user profile for personalizing language model outputs. |
| Outcome: | The proposed approach is effective for a set of zero-shot and fine-tuned language models and highlights the impact of personalization in various natural language tasks. |
Beyond Factual Accuracy: Evaluating Coverage of Diverse Factual Information in Long-form Text Generation (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing evaluation frameworks for large language models focus on isolated aspects like * Equal contribution. |
| Approach: | They evaluate ICAT, an evaluation framework for measuring coverage of diverse factual information in long-form text generation. |
| Outcome: | The evaluation framework is based on three implementations with different assumptions on availability of aspects and alignment method. |
ExPerT: Effective and Explainable Evaluation of Personalized Long-Form Text Generation (2025.findings-acl)
Copied to clipboard
| Challenge: | Evaluating personalized text generated by large language models is challenging, as only the LLM user, i.e. prompt author, can reliably assess the output. |
| Approach: | They propose an explainable reference-based evaluation framework that leverages an LLM to extract atomic aspects and their evidences from the generated and reference texts, match the aspects, and evaluate their alignment based on content and writing style. |
| Outcome: | The proposed framework achieves a 7.2% improvement in alignment with human judgments compared to the state-of-the-art evaluation methods. |
You can’t pick your neighbors, or can you? When and How to Rely on Retrieval in the kNN-LM (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Recent studies have shown that retrieval-enhanced language models can improve perplexity by combining text from large external datastores with a k-nearest neighbors model. |
| Approach: | They propose a retrieval-enhanced language model that interpolates existing LMs with a k-nearest neighbors model and requires no additional training. |
| Outcome: | The proposed model improves on two English language modeling datasets and shows that it is most effective when items have high semantic similarity with the query. |