CORTEXA
← Browse
arxivcs.LG2026-06-29

Building Multi-Task Agentic LLMs via Two-Phase Distillation

Huaijie Wang, Shusheng Xu, Yi Wu, Kaifeng Lyu

A key step toward artificial general intelligence is to train models that can perform multiple tasks. In this paper, we study how to build such models by first training separate RL experts for individual tasks and then consolidating them via distillation, as an alternative to directly training a single model on mixed tasks. We show that off-policy distillation degrades in multi-task settings due to the mode-covering nature of forward KL: aggregating data from multiple tasks introduces a large number of behavioral modes that can exceed the student's capacity, forcing it to average across behaviors and leading to degraded performance. In contrast, on-policy distillation is mode-seeking but requires strong initialization. Inspired by these observations, we propose a two-phase approach: off-policy distillation followed by on-policy refinement. Evaluation across conversational agents and text-based games confirms that this two-phase approach matches single-task RL expert performance for each individual task, whereas off-policy or on-policy distillation alone fails to match this performance.

View free PDFSource page

Related papers

arxivcs.LGcs.AI2026-07-09

Open-ended Multi-agent Autocurricula via Visual Inspection of Policies with Multi-modal LLMs

Lorenzo Pantè, Andrea Fanti, Roberto Capobianco

Open-ended curricula in Reinforcement Learning (RL) aim to train generally-capable agents by identifying tasks that facilitate learning increasingly complex skills. A major challenge when designing such curricula is assessing task difficulty relative to the agent's current learni…

View free PDFSource page
arxivcs.LGcs.ITeess.SPstat.ML2026-07-23

Semantic-Aware Task Clustering for Constructive and Cooperative Multi-Tasking

Ahmad Halimi Razlighi, Maximilian H. V. Tillmann, Edgar Beck, Bho Matthiesen, Armin Dekorsy

Cooperative multi-task semantic communication (CMT-SemCom) improves task execution performance by leveraging shared representations. However, as we demonstrated in [1], cooperative multi-tasking can be either constructive or destructive, depending on the semantic relationships am…

View free PDFSource page
arxivcs.LGcs.AIcs.IT2026-07-17

Capacity and Redundancy Trade-offs in Multi-Task Learning

Asif Khan

In multi-task learning (MTL) negative transfer is often considered as an optimization artifact, but it can also be viewed as a consequence of limited shared capacity and weak task redundancy. We investigate this effect through a Capacity--Redundancy (CR) identity that decomposes…

View free PDFSource page
arxivcs.LG2026-07-17

Leakage-Robust Evaluation and Data-Scale Sensitivity of Attention-Enhanced Multi-Task Learning for Joint Fault Diagnosis and Remaining Useful Life Estimation

Md Mahamudur Rahaman Shamim, Md. Nuruzzaman, Zannatul Ferdus, Md Rajib Ahmed, Abieer Nwshad Anward, Mohammad Tooneer, et al.

Multi-task deep learning models that jointly perform fault classification and remaining useful life (RUL) regression are increasingly used in predictive maintenance, yet reported performance can be strongly affected by how sliding-window sequences are split into training and test…

View free PDFSource page
arxivcs.LGcs.AI2026-07-08

Entropy Pacing Policy Optimization for Multi-Task Agentic Reinforcement Learning

Zetian Hu, Shunyu Liu, Junjie Zhang, Yongcheng Jing, Ting-En Lin, Yongbin Li, et al.

Recent breakthroughs of Reinforcement Learning (RL) have highlighted its potential for complex agentic Large Language Model (LLM) tasks. However, existing efforts largely focus on single-task settings, whereas real-world deployment necessitates a generalist agent capable of solvi…

View free PDFSource page