Skills on the Fly: Test-Time Adaptive Skill Synthesis for LLM Agents

• URL: http://arxiv.org/abs/2605.16986v1
• Date: Sat, 16 May 2026 13:14:15 GMT
• Title: Skills on the Fly: Test-Time Adaptive Skill Synthesis for LLM Agents
• Abstract summary: We propose emphSkillTTA, a Test-Time Adaptive Skill Synthesis method.<n>It retrieves a small set of training trajectories relevant to the current task and synthesizes them into a temporary, task-specific textual skill.<n>We evaluate the method on SpreadsheetBench, ALFWorld, and BigCodeBench.
• Score: 33.57540444018471
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: LLM agents benefit from reusable skills, yet test-time tasks often require guidance more specific than a static skill library can provide. We propose \emph{SkillTTA}, a Test-Time Adaptive Skill Synthesis method that retrieves a small set of training trajectories relevant to the current task and synthesizes them into a temporary, task-specific textual skill. The solver model is kept fixed, so adaptation happens entirely through generated context rather than parameter updates. We evaluate the method on SpreadsheetBench, ALFWorld, and BigCodeBench. Compared with static trajectory-to-skill synthesis using GPT-5.5, task-specific skills improve SpreadsheetBench Pass@1 from 0.397 to 0.505 and BigCodeBench Pass@1 from 0.517 to 0.651. On ALFWorld, the method matches a heavier memory-learning baseline within four points of success rate while producing the shortest successful trajectories among reported methods. Ablations on SpreadsheetBench further show that synthesized skills outperform raw trajectory prompting, that top-$k$ retrieval should stay small, and that failed trajectories are especially useful because they expose recurring evaluator-facing mistakes.

Related papers

• Beyond Solvability: Task Learnability as a Static Prior for LLM RL Post-Training [52.78248352851176]
  Reinforcement learning (RL) has become a central post-training paradigm for eliciting reasoning capabilities in large language models.<n>We study this residual axis as task learnability: a regime-conditional measure of expected positive response to continued training.<n>We propose TrajVal, a lightweight probe-based estimator that approximates per-task learnability from a short probe run and two endpoint evaluations.
  arXiv  Detail & Related papers  (2026-08-10T07:43:05Z)
• SkillCAT: Contrastive Assessment and Topology-Aware Skill Self-Evolution for LLM Agents [42.30624901249764]
  Current pipelines learn from one trajectory per task, merge candidate skill patches before checking them, and load the full skill corpus before inference.<n>We propose SkillCAT, a training-free framework that separates this process into three stages.<n>We evaluate SkillCAT on common agent benchmarks, including SpreadsheetBench, WikiTableQuestions, and DocVQA.
  arXiv  Detail & Related papers  (2026-06-11T13:12:10Z)
• SKILL0: In-Context Agentic Reinforcement Learning for Skill Internalization [61.37478254738943]
  Inference-time skill augmentation is fundamentally limited.<n>We introduce SKILL0, an in-context reinforcement learning framework designed for skill internalization.<n>SKILL0 achieves substantial improvements over the standard RL baseline.
  arXiv  Detail & Related papers  (2026-04-02T17:03:05Z)
• MetaClaw: Just Talk -- An Agent That Meta-Learns and Evolves in the Wild [74.7263562191605]
  Large language model (LLM) agents are increasingly used for complex tasks.<n>Existing methods either store raw trajectories without distilling knowledge, maintain static skill libraries, or require disruptive downtime for retraining.<n>We present MetaClaw, a continual meta-learning framework that jointly evolves a base LLM policy and a library of reusable behavioral skills.
  arXiv  Detail & Related papers  (2026-03-17T22:30:30Z)
• CoT-based Synthesizer: Enhancing LLM Performance through Answer Synthesis [31.953858122298517]
  We propose a novel inference scaling strategy, CoT-based Synthesizer.<n>It synthesizes superior answers by analyzing complementary information from multiple candidate responses.<n>We show that our method significantly enhances performance, with gains of 11.8% for Llama3-8B and 10.3% for GPT-4o.
  arXiv  Detail & Related papers  (2025-01-03T06:50:06Z)
• TL-Training: A Task-Feature-Based Framework for Training Large Language Models in Tool Use [72.32614703504122]
  Large language models (LLMs) achieve remarkable advancements by leveraging tools to interact with environments.<n>Standard supervised fine-tuning approach, which relies on large-scale datasets, often overlooks task-specific characteristics in tool use.<n>We propose TL-Training, a task-feature-based framework that mitigates the effects of suboptimal training data.
  arXiv  Detail & Related papers  (2024-12-20T02:21:36Z)
• The Surprising Effectiveness of Test-Time Training for Few-Shot Learning [59.309477460893916]
  Language models (LMs) have shown impressive performance on tasks within their training distribution, but often struggle with structurally novel tasks.<n>We investigate the effectiveness of test-time training (TTT) as a mechanism for improving LMs' reasoning and few-shot learning capabilities.<n>Our findings highlight the limitations of in-context learning for novel tasks and demonstrate the potential of test-time training to enhance language model adaptability.
  arXiv  Detail & Related papers  (2024-11-11T18:59:45Z)
• Improving Representational Continuity via Continued Pretraining [76.29171039601948]
  Transfer learning community (LP-FT) outperforms naive training and other continual learning methods. LP-FT also reduces forgetting in a real world satellite remote sensing dataset (FMoW) variant of LP-FT gets state-of-the-art accuracies on an NLP continual learning benchmark.
  arXiv  Detail & Related papers  (2023-02-26T10:39:38Z)
• Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learning [81.3514358542452]
  Few-shot in-context learning (ICL) incurs substantial computational, memory, and storage costs because it involves processing all of the training examples every time a prediction is made. parameter-efficient fine-tuning offers an alternative paradigm where a small set of parameters are trained to enable a model to perform the new task. In this paper, we rigorously compare few-shot ICL and parameter-efficient fine-tuning and demonstrate that the latter offers better accuracy as well as dramatically lower computational costs.
  arXiv  Detail & Related papers  (2022-05-11T17:10:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.