paper-with-me

홈 › Papers

TELL-TALE: Task Efficient LLMs with Task Aware Layer Elimination

2025-10-26 · Omar Naim, Krish Sharma, Niyar R Barman, Nicholas Asher arxiv

Large Language Models (LLMs) typically come with a fixed architecture, despite growing evidence that not all layers contribute equally to every downstream task. We introduce TALE (Task-Aware Layer Elimination), an inference-time method that improves task performance by selectively removing layers that are irrelevant or detrimental for a given task. TALE optimizes task-specific performance, yielding a task-optimized architecture without retraining. Across 9 tasks and 5 model families, under both zero-shot and few-shot settings, TALE consistently matches or surpasses baseline performance while simultaneously reducing computational costs. TALE also synergizes with fine-tuning, leading to further performance improvements. Computing TALE for a new task requires modest resources, making it a practical and deployable solution for task-specialized LLM inference.

📄 PDF Abstract BibTeX arXiv:2510.22767

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Do Large Language Models Know Folktales? A Case Study of Yokai in Japanese Folktales

2025-06-04 · Ayuto Tsutsumi, Yuu Jinnai

Although Large Language Models (LLMs) have demonstrated strong language understanding and generation abilities across various languages, their cultural knowledge is often limited to English-speaking communities, which ca…

Multiple-choice

Enhancing Talent Search Ranking with Role-Aware Expert Mixtures and LLM-based Fine-Grained Job Descriptions

2025-10-05 · Jihang Li, Bing Xu, Zulong Chen, Chuanfei Xu 외 arxiv

Talent search is a cornerstone of modern recruitment systems, yet existing approaches often struggle to capture nuanced job-specific preferences, model recruiter behavior at a fine-grained level, and mitigate noise from …

Multi-Task Learning

LLM-Guided Task- and Affordance-Level Exploration in Reinforcement Learning

2025-09-20 · Jelle Luijkx, Runyu Ma, Zlatan Ajanović, Jens Kober arxiv

Reinforcement learning (RL) is a promising approach for robotic manipulation, but it can suffer from low sample efficiency and requires extensive exploration of large state-action spaces. Recent methods leverage the comm…

Reinforcement Learning

Freshness-Aware Prioritized Experience Replay for LLM/VLM Reinforcement Learning

2026-04-18 · Weiyu Ma, Yongcheng Zeng, Yan Song, Xinyu Cui 외 arxiv

Reinforcement Learning (RL) has achieved impressive success in post-training Large Language Models (LLMs) and Vision-Language Models (VLMs), with on-policy algorithms such as PPO, GRPO, and REINFORCE++ serving as the dom…

Reinforcement Learning

PhoniTale: Phonologically Grounded Mnemonic Generation for Typologically Distant Language Pairs

2025-07-07 · Sana Kang, Myeongseok Gwon, Su Young Kwon, Jaewook Lee 외 arxiv

Vocabulary acquisition poses a significant challenge for second-language (L2) learners, especially when learning typologically distant languages such as English and Korean, where phonological and structural mismatches co…