Papers by Devleena Das
DEFT-UCS: Data Efficient Fine-Tuning for Pre-Trained Language Models via Unsupervised Core-Set Selection for Text-Editing (2024.emnlp-main)
Copied to clipboard
| Challenge: | Recent advances in language modelling have led to the availability of many pre-trained language models (PLMs); however, how much data is needed to fine-tune PLMs for downstream tasks? |
| Approach: | They propose a data-efficient fine-tuning framework that leverages unsupervised core-set selection to identify a smaller, representative dataset to fine- tune PLMs for text-editing tasks. |
| Outcome: | The proposed framework is as accurate as CoEDIT across eight different datasets consisting of six different editing tasks, while finetuning on 70% less data. |
Recover-LoRA: Data-Free Accuracy Recovery of Degraded Language Models via Low-Rank Adaptation (2025.emnlp-industry)
Copied to clipboard
| Challenge: | Inference optimizations such as quantization, pruning, format and datatype conversion, model export, and serialization can lead to functional degradations in language model task performance. |
| Approach: | They propose a lightweight and dataset-agnostic method to recover model accuracies from quantization, pruning, format and datatype conversion, model export, and serialization errors. |
| Outcome: | The proposed method recovers model accuracies by 5-17% on MHA and GQA models. |
KV Pareto: Systems-Level Optimization of KV Cache and Model Compression for Long Context Inference (2026.eacl-industry)
Copied to clipboard
| Challenge: | Long-context Large Language Models (LLMs) face significant memory bottlenecks due to the linear growth of key-value (KV) cache with sequence length. |
| Approach: | They propose a framework that maps the trade-off frontier between total memory consumption and task accuracy across three complementary optimization techniques. |
| Outcome: | The proposed model-specific configurations achieve 68-78% total memory reduction with minimal (1-3%) accuracy degradation on long-context tasks. |