Papers by Salman Avestimehr

12 papers
Ethos: Rectifying Language Models in Orthogonal Parameter Space (2024.findings-naacl)

Copied to clipboard

Challenge: Language models (LMs) generate toxic, biased content and reveal private training records.
Approach: They propose an efficient approach that rectifies LMs to mitigate toxicity and bias . Ethos distinguishes general beneficial and undesired knowledge when reconstructing task vectors .
Outcome: The proposed approach mitigates toxicity and bias in outputs and avoids privacy leakage.
ScaleLLM: A Resource-Frugal LLM Serving Framework by Optimizing End-to-End Efficiency (2024.emnlp-industry)

Copied to clipboard

Challenge: Large language models (LLMs) are widely used in commercial applications . low latency is crucial due to system latency, query concurrency, and computational resources constraints.
Approach: They propose a system that can be resource-efficiently served by addressing bottlenecks beyond LLM inference . they propose 4.3 speed up over vLLM and 1.5 higher throughput .
Outcome: The proposed system outperforms state-of-the-arts with 1.5 higher throughput . it achieves 4.3 speed up with 64 concurrent requests on Mixtral 8x7B .
TensorOpera Router: A Multi-Model Router for Efficient LLM Inference (2024.emnlp-industry)

Copied to clipboard

Challenge: Large Language Models (LLMs) have demonstrated remarkable performance across a diverse set of domain-specific tasks.
Approach: They propose a non-monolithic LLM querying system that seamlessly integrates various LLM experts into a single query interface and dynamically routes incoming queries to the most high-performant expert based on query’s requirements.
Outcome: The proposed model improves query efficiency by 40% and costs by 30% while maintaining or enhancing model performance by 10%.
Do Not Design, Learn: A Trainable Scoring Function for Uncertainty Estimation in Generative LLMs (2025.findings-naacl)

Copied to clipboard

Challenge: Existing methods for probability-based UE are limited by their inability to handle biased probabilities and complex semantic dependencies between tokens.
Approach: They propose a learning-based scoring function that captures complex dependencies between tokens and probabilities and produces more reliable responses.
Outcome: The proposed function outperforms existing scoring functions in question-answering and arithmetical reasoning tasks with different datasets.
TruthTorchLM: A Comprehensive Library for Predicting Truthfulness in LLM Outputs (2025.emnlp-demos)

Copied to clipboard

Challenge: Generative Large Language Models (LLMs) produce untruthful outputs, referred to as hallucinations, which are often referred as false positives.
Approach: They propose an open-source Python library with over 30 truthfulness prediction methods.
Outcome: The proposed methods span diverse trade-offs in computational cost, access level, grounding document requirements, and supervision type (self-supervised or supervised).
FedNLP: Benchmarking Federated Learning Methods for Natural Language Processing Tasks (2022.findings-naacl)

Copied to clipboard

Challenge: Increasing concerns and regulations about data privacy necessitate the study of privacy-preserving, decentralized learning methods for natural language processing tasks.
Approach: They propose a framework for evaluating federated learning methods on four different tasks . they propose federation between Transformer-based language models and FL methods .
Outcome: The proposed framework compares FL methods on four different tasks under non-IID partitioning strategies.
Creating a Lens of Chinese Culture: A Multimodal Dataset for Chinese Pun Rebus Art Understanding (2025.findings-acl)

Copied to clipboard

Challenge: a new study examines the performance of large vision-language models in understanding art . the Pun Rebus Art Dataset is a multimodal dataset for art understanding rooted in traditional Chinese culture .
Approach: They propose a multimodal dataset for art understanding deeply rooted in traditional Chinese culture . they aim to facilitate the development of VLMs that can better understand culturally specific content .
Outcome: The proposed dataset shows that state-of-the-art VLMs struggle with these tasks . the data will facilitate the development of VLM models that can better understand culturally specific content .
Reconsidering LLM Uncertainty Estimation Methods in the Wild (2025.acl-long)

Copied to clipboard

Challenge: Existing studies evaluate UE methods in short-form QA settings, but real-world deployment presents several challenges.
Approach: They examine UE methods' sensitivity to decision threshold selection and their robustness to query transformations such as typos and adversarial prompts.
Outcome: The proposed methods exhibit robustness against typos, adversarial prompts, and prior chat history, and are highly susceptible to adversarials.
Revisiting OPRO: The Limitations of Small-Scale LLMs as Optimizers (2024.findings-acl)

Copied to clipboard

Challenge: Recent studies aim to enhance the efficacy of Large Language Models (LLMs) through strategic prompting.
Approach: They propose to revisit the optimization by prompting approach for small-scale LLMs . they suggest future prompting engineering to consider both model capabilities and computational costs .
Outcome: The proposed approach shows limited effectiveness in small-scale LLMs, with limited inference capabilities constraining optimization ability.
MobiZO: Enabling Efficient LLM Fine-Tuning at the Edge via Inference Engines (2025.emnlp-main)

Copied to clipboard

Challenge: Large Language Models (LLMs) are currently pre-trained and fine-tuned on large cloud servers . fine-timing on resource-constrained edge devices presents significant memory and computational demands .
Approach: They propose a resource-efficient fine-tuning framework for LLMs specifically designed for edge devices.
Outcome: Experiments show that MobiZO achieves substantial runtime speedups and memory savings while improving fine-tuning accuracy.
Federated Learning with Noisy User Feedback (2022.naacl-main)

Copied to clipboard

Challenge: Artificial Intelligence (AI) and Machine Learning (ML) systems are becoming more popular and are causing concerns over user privacy.
Approach: They propose a method for training ML models using positive and negative user feedback and a framework to extract labels on edge to make FL viable.
Outcome: The proposed method improves significantly over a self-training baseline, achieving performance closer to models trained with full supervision.
MARS: Meaning-Aware Response Scoring for Uncertainty Estimation in Generative LLMs (2024.acl-long)

Copied to clipboard

Challenge: Generative Large Language Models (LLMs) are widely utilized for their excellence in various tasks. however, their tendency to produce inaccurate or misleading outputs poses a potential risk.
Approach: They propose a new scoring function that considers the semantic contribution of each token in the generated sequence in the context of the question.
Outcome: The proposed scoring function improves UE performance on a medical QA dataset.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations