paper-with-me

홈 › Papers

Search, Do not Guess: Teaching Small Language Models to Be Effective Search Agents

2026-04-06 · Yizhou Liu, Qi Sun, Yulin Chen, Siyue Zhang, Chen Zhao arxiv

Agents equipped with search tools have emerged as effective solutions for knowledge-intensive tasks. While Large Language Models (LLMs) exhibit strong reasoning capabilities, their high computational cost limits practical deployment for search agents. Consequently, recent work has focused on distilling agentic behaviors from LLMs into Small Language Models (SLMs). Through comprehensive evaluation on complex multi-hop reasoning tasks, we find that despite possessing less parametric knowledge, SLMs invoke search tools less frequently and are more prone to hallucinations. To address this issue, we propose \policy, a lightweight fine-tuning approach that explicitly trains SLMs to reliably retrieve and generate answers grounded in retrieved evidence. Compared to agent distillation from LLMs, our approach improves performance by 17.3 scores on Bamboogle and 15.3 scores on HotpotQA, achieving LLM-level results across benchmarks. Our further analysis reveals that adaptive search strategies in SLMs often degrade performance, highlighting the necessity of consistent search behavior for reliable reasoning.

📄 PDF Abstract BibTeX arXiv:2604.04651

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Teaching and Learning under Deductive Errors

2026-05-13 · Jan Arne Telle, Brigt Håvardstun, Jose Hernandez-Orallo arxiv

Most models of machine teaching and learning assume the learner makes no errors in its internal deductive inference. However, humans and large language models in few-shot learning regimes are two important examples of le…

Few-Shot Learning

Teaching Transformers to Solve Combinatorial Problems through Efficient Trial & Error

2025-09-26 · Panagiotis Giannoulis, Yorgos Pantis, Christos Tzamos arxiv

Despite their proficiency in various language tasks, Large Language Models (LLMs) struggle with combinatorial problems like Satisfiability, Traveling Salesman Problem, or even basic arithmetic. We address this gap throug…

Stochastic Optimization

From Blind Guess to Informed Judgment: Teaching LLMs to Evaluate Materials by Building Knowledge-Augmented Preference Signals

2026-05-28 · Yeyong Yu, Wenya Hu, Xing Wu, Quan Qian arxiv

As candidate generation and high-throughput experimentation advance, the primary bottleneck in materials discovery is shifting from property prediction to making reliable evaluations among massive candidate sets. We prop…

ArchiGuesser -- AI Art Architecture Educational Game

2023-12-14 · Joern Ploennigs, Markus Berger, Eva Carnein

The use of generative AI in education is a controversial topic. Current technology offers the potential to create educational content from text, speech, to images based on simple input prompts. This can enhance productiv…

Board GamesDiversityImage Generation

When Redundancy Matters: Machine Teaching of Representations

2024-01-23 · Cèsar Ferri, Dario Garigliotti, Brigt Arve Toppe Håvardstun, Josè Hernández-Orallo 외

In traditional machine teaching, a teacher wants to teach a concept to a learner, by means of a finite set of examples, the witness set. But concepts can have many equivalent representations. This redundancy strongly aff…