Papers by Daniel Schwartz
Zero-knowledge LLM hallucination detection and mitigation through fine-grained cross-model consistency (2025.emnlp-industry)
Copied to clipboard
| Challenge: | Existing methods for hallucination management fail to integrate both detection and mitigation without external knowledge sources. |
| Approach: | They propose a black-box framework that leverages fine-grained cross-model consistency to detect and mitigate hallucinations in LLM outputs without external knowledge sources. |
| Outcome: | The proposed framework improves hallucination detection scores by 6-39% on a FELM dataset . it achieves 9 percentage points improvement in answer accuracy on the GPQA-diamond dataset compared to existing approaches . |
How Much Does Attention Actually Attend? Questioning the Importance of Attention in Pretrained Transformers (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Pretrained language models use the attention mechanism to contextualize input inputs . but, we find that it is not as important as thought for pretrained models . |
| Approach: | They propose a probing method that replaces input-dependent attention matrices with constant ones. |
| Outcome: | The proposed method improves performance of pretrained language models without input-dependent attention. |
The Remarkable Benefit of User-Level Aggregation for Lexical-based Population-Level Predictions (D18-1)
Copied to clipboard
Salvatore Giorgi, Daniel Preoţiuc-Pietro, Anneke Buffone, Daniel Rieman, Lyle Ungar, H. Andrew Schwartz
| Challenge: | Social media data is often aggregated without regard to users in the Twitter populations of each community. |
| Approach: | They propose to use Twitter language to build community-level models using Twitter language aggregated by users. |
| Outcome: | The proposed method improves on four county-level tasks spanning demographic, health, and psychological outcomes over the standard approach of aggregating all tweets. |
Finding the SWEET Spot: Analysis and Improvement of Adaptive Inference in Low Resource Settings (2023.acl-long)
Copied to clipboard
| Challenge: | Pre-trained Transformer-based language models such as BERT, DeBERTa, and GPT3 have become the go-to tool in NLP. |
| Approach: | They propose an Early-Exit fine-tuning method that assigns each classifier its own set of unique model weights, not updated by other classifiers. |
| Outcome: | The proposed method outperforms Early-Exit and Multi-Model at fast speeds while maintaining comparable scores to Early- Exit at slow speeds. |
Extracting a Knowledge Base of Mechanisms from COVID-19 Papers (2021.naacl-main)
Copied to clipboard
Tom Hope, Aida Amini, David Wadden, Madeleine van Zuylen, Sravanthi Parasa, Eric Horvitz, Daniel Weld, Roy Schwartz, Hannaneh Hajishirzi
| Challenge: | COVID-19 has spawned a diverse body of scientific literature that is challenging to navigate . researchers are using automated tools to help find useful knowledge . |
| Approach: | They develop a schema to extract mechanism relations from scientific papers . their search engine, dataset and code are publicly available . |
| Outcome: | The proposed schema outperforms PubMed search in clinical trials. |
Graph of Attacks with Pruning: Optimizing Stealthy Jailbreak Prompt Generation for Enhanced LLM Content Moderation (2025.emnlp-industry)
Copied to clipboard
| Challenge: | Existing jailbreaking methods create adversarial prompts to bypass LLM safeguards. |
| Approach: | They propose a framework for generating stealthy jailbreak prompts that enables knowledge sharing across attack paths. |
| Outcome: | The proposed framework outperforms state-of-the-art methods for attacking both open and closed LLMs with attack success rates of >96%. |