Papers by Ahmad Beirami
Improving Neutral Point-of-View Generation with Data- and Parameter-Efficient RL (2025.emnlp-main)
Copied to clipboard
Jessica Hoffmann, Christiane Ahlheim, Zac Yu, Aria Walfrand, Jarvis Jin, Marie Tano, Ahmad Beirami, Erin MacMurray van Liemt, Nithum Thain, Hakim Sidahmed, Lucas Dixon
| Challenge: | Parameter-efficient reinforcement learning (PE-RL) is a highly effective training regime to improve large language models’ ability to answer queries on sensitive topics with a Neutral Point of View (NPOV). |
| Approach: | They propose to use parameter-efficient reinforcement learning to train large language models to answer queries with a Neutral Point of View (NPOV) This is compared to the strongest baseline, LoRA finetuning, SFT and RLHF. |
| Outcome: | The proposed training regime improves on NPOV quality and scores higher on features identified by linguists as key to separating good answers from the best answers. |
Database Search Results Disambiguation for Task-Oriented Dialog Systems (2022.naacl-main)
Copied to clipboard
Kun Qian, Satwik Kottur, Ahmad Beirami, Shahin Shayandeh, Paul Crook, Alborz Geramifard, Zhou Yu, Chinnadhurai Sankar
| Challenge: | Task-oriented dialog systems can't handle multiplesearch results when querying a database due to the lack of such scenarios in existing datasets. |
| Approach: | They propose a task that focuses on disambiguating database search results by synthetically generating turns through a pre-defined grammar and collecting human paraphrases for a subset. |
| Outcome: | The proposed task improves performance on DSR-disambiguation even in the absence of in-domain data, suggesting it can be learned as a universal dialog skill. |
Reuse Your Rewards: Reward Model Transfer for Zero-Shot Cross-Lingual Alignment (2024.emnlp-main)
Copied to clipboard
| Challenge: | Multilingual human preference data are difficult to obtain at scale, making it challenging to extend this framework to diverse languages. |
| Approach: | They propose a method where a reward model is trained on preference data in one source language and applied to other target languages. |
| Outcome: | The proposed approach is effective under comprehensive evaluation settings, including human evaluation. |
Know Thy Strengths: Comprehensive Dialogue State Tracking Diagnostics (2022.findings-emnlp)
Copied to clipboard
Hyundong Cho, Chinnadhurai Sankar, Christopher Lin, Kaushik Sadagopan, Shahin Shayandeh, Asli Celikyilmaz, Jonathan May, Ahmad Beirami
| Challenge: | Recent studies have revealed the vulnerability of dialogue state tracking models to distributional shifts, resulting in poor performance. |
| Approach: | They present a toolkit for standardized and comprehensive dialogue state tracking diagnoses that provides a richer summary of strengths and weaknesses. |
| Outcome: | The proposed toolkit shows that different classes of DST models have clear strengths and weaknesses, while generation models are more promising for handling language variety and span-based classification models are robust to unseen entities. |
Robust Conversational Agents against Imperceptible Toxicity Triggers (2022.naacl-main)
Copied to clipboard
| Challenge: | Existing work to generate adversarial attacks is costly and not scalable . despite the abundance of research in this area, little attention has been given to adversarials . |
| Approach: | They propose an adversarial attack mechanism that mitigates toxic language generation . they propose a defense mechanism that is scalable and can be generalized . |
| Outcome: | The proposed defense is effective at avoiding toxic language generation even against imperceptible toxicity triggers while preserving conversational flow. |
Resource Constrained Dialog Policy Learning Via Differentiable Inductive Logic Programming (2020.coling-main)
Copied to clipboard
| Challenge: | Existing methods for dialog policy learning have limited data collection and data analysis. |
| Approach: | They introduce dialog policy learning via differentiable inductive logic on SimDial and MultiWoZ to address resource constrained dialog policy. |
| Outcome: | The proposed method is 100x more data efficient than state-of-the-art neural approaches on MultiWoZ while achieving similar performance metrics. |
Situated and Interactive Multimodal Conversations (2020.coling-main)
Copied to clipboard
Seungwhan Moon, Satwik Kottur, Paul Crook, Ankita De, Shivani Poddar, Theodore Levin, David Whitney, Daniel Difranco, Ahmad Beirami, Eunjoon Cho, Rajen Subba, Alborz Geramifard
| Challenge: | Situated Interactive MultiModal Conversations (SIMMC) is a new direction for virtual assistants that handle multimodal inputs and perform multimodal actions. |
| Approach: | They propose to use Situated Interactive MultiModal Conversations (SIMMC) to train agents to take multimodal actions grounded in a co-evolving multimodal context. |
| Outcome: | The proposed model will be made publicly available. |
DVD: A Diagnostic Dataset for Multi-step Reasoning in Video Grounded Dialogue (2021.acl-long)
Copied to clipboard
| Challenge: | Existing benchmarks do not have enough annotations to analyze video-grounded dialogue systems and understand their capabilities and limitations in isolation. |
| Approach: | They present a Diagnostic Dataset for Video-grounded dialogue with minimal biases and detailed annotations for the different types of reasoning over the spatio-temporal space of video. |
| Outcome: | The proposed system is based on 11k CATER synthetic videos and contains 10 instances of 10-round dialogues for each video. |
Improving Diversity of Demographic Representation in Large Language Models via Collective-Critiques and Self-Voting (2023.emnlp-main)
Copied to clipboard
Preethi Lahoti, Nicholas Blumm, Xiao Ma, Raghavendra Kotikalapudi, Sahitya Potluri, Qijun Tan, Hansa Srinivasan, Ben Packer, Ahmad Beirami, Alex Beutel, Jilin Chen
| Challenge: | Existing studies on diversity in large language models focus on the understudied class of fairness and inclusion concern in LLMs. |
| Approach: | They propose a technique to measure diversity in generated responses along people and culture axes by collective-critique and self-voting. |
| Outcome: | The proposed approach outperforms baseline methods and human evaluations with human and automated evaluations. |
Gradient-Based Language Model Red Teaming (2024.eacl-long)
Copied to clipboard
| Challenge: | generative language models generate unsafe responses by producing adversarial prompts . red teaming is labor-intensive and difficult to scale when done by humans. |
| Approach: | They propose a red teaming method that generates diverse prompts that are likely to cause an LM to generate unsafe responses. |
| Outcome: | The proposed method is more effective at finding prompts that trigger an LM to generate unsafe responses than a strong reinforcement learning-based red teaming approach. |