Papers by Ravi Tandon
Can Multiple Responses from an LLM Reveal the Sources of Its Uncertainty? (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Large language models can produce unreliable or misleading outputs, posing challenges for real-world applications. |
| Approach: | They employ an auxiliary LLM to analyze the patterns of disagreement among LLMs . they validate their framework on AmbigQA, OpenBookQA, and MMLU-Pro . |
| Outcome: | The proposed model can be used to diagnose uncertainty sources in a model with an auxiliary model. |
STAMP: Selective Task-Aware Mechanism for Text Privacy (2026.eacl-long)
Copied to clipboard
| Challenge: | Experimental evaluations on SQuAD, Yelp, and AG News datasets demonstrate that STAMP achieves superior privacy–utility trade-offs across varying per-token privacy budgets. |
| Approach: | They propose a new framework for task-aware text privatization that selectively allocates privacy budgets across tokens by jointly considering (i) each token’s importance to the downstream task and (ii) its privacy sensitivity. |
| Outcome: | The proposed framework achieves superior privacy–utility trade-offs on SQuAD, Yelp, and AG News datasets. |