activity
20242026
collaborators

8 papers

cs.CL2026

Two-Stage Prompt Optimization for Few-Shot Relation Extraction: From Reasoning-Guided Search to Gradient-Guided Refinement

Aunabil Chakma, Mihai Surdeanu, Eduardo Blanco

Automatic prompt optimization is still underexplored for episodic few-shot relation extraction with smaller language models. We propose a two-stage framework that combines reasonin…

cs.CL2026

Structured Semantic Information Helps Retrieve Better Examples for In-Context Learning Applied to Few-Shot Relation Extraction

Aunabil Chakma, Mihai Surdeanu, Eduardo Blanco

This paper presents several strategies to automatically obtain additional examples for in-context learning, effectively transforming relation extraction from a 1-shot to a few-shot…

cs.CL2026

Bridging the Long-Tail Gap: Robust Retrieval-Augmented Relation Completion via Multi-Stage Paraphrase Infusion

Fahmida Alam, Mihai Surdeanu, Ellen Riloff

Large language models (LLMs) struggle with relation completion (RC), both with and without retrieval-augmented generation (RAG), particularly when the required information is rare…

cs.LG2026

The Alchemy of Thought: Understanding In-Context Learning Through Supervised Classification

Harshita Narnoli, Mihai Surdeanu

In-context learning (ICL) has become a prominent paradigm to rapidly customize LLMs to new tasks without fine-tuning. However, despite the empirical evidence of its usefulness, we…

cs.CL2025

Can LLMs Judge Debates? Evaluating Non-Linear Reasoning via Argumentation Theory Semantics

Reza Sanayei, Srdjan Vesic, Eduardo Blanco +1

Large Language Models (LLMs) excel at linear reasoning tasks but remain underexplored on non-linear structures such as those found in natural debates, which are best expressed as a…

cs.CL2025

Towards Compute-Optimal Many-Shot In-Context Learning

Shahriar Golchin, Yanfei Chen, Rujun Han +7

Long-context large language models (LLMs) are able to process inputs containing up to several million tokens. In the scope of in-context learning (ICL), this translates into using…