Papers by Yash Kumar Atri
Evaluating Temporal Consistency in Multi-Turn Language Models (2026.acl-long)
Copied to clipboard
| Challenge: | Language models are increasingly deployed in interactive settings where users reason about facts over time . we study temporal scope stability, the ability to preserve, override, or transfer time-scoped factual context across dialogue turns. |
| Approach: | They propose a diagnostic benchmark to isolate temporal scope stability in controlled multi-turn interactions. |
| Outcome: | The proposed model can preserve, override, or transfer time-scoped factual context across dialogue turns. |
Lifelong Model Editing with Graph-Based External Memory (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for post-training model editing suffer from overfitting and catastrophic forgetting. |
| Approach: | They propose a framework that leverages hyperbolic geometry and graph neural networks for precise and stable model edits. |
| Outcome: | Experiments on CounterFact, CounterFACT+, and MQuAKE with GPT2-XL and GPT-J show that HYPE significantly enhances edit stability, factual accuracy, and multi-hop reasoning. |