Trendoday

Agentic On-Policy Distillation Implicitly Reshapes LLMs' ReAct Patterns

On-policy distillation (OPD) transfers specialized capabilities to language-model agents, but its cross-domain effects on interaction remain unclear. We investigate how OPD reshapes ReAct interaction

  • Upvotes7
–Rank now
#18Peak rank
2Days on list
121Score · 7d
Nov 18, 2026First seen
Rank · 7 days

About this paper

On-policy distillation can improve or degrade language-model agent interaction across domains, and selectively retaining teacher-preferred positions mitigates negative transfer at a cost to source-domain learning.

On-policy distillation (OPD) transfers specialized capabilities to language-model agents, but its cross-domain effects on interaction remain unclear. We investigate how OPD reshapes ReAct interaction patterns across mathematics, ALFWorld, and τ^2-bench, focusing on reasoning, actions, and observation acquisition. General-reasoning supervision reduces mathematical tool attempts from 70.4% to 14.4% while increasing accuracy from 16.7% to 38.0%; code-interpreter supervision improves household actions and retail queries. In contrast, ALFWorld supervision reduces τ^2-bench accuracy from 41.5% to 6.6%, accompanied by frequent tool calls that violate user-information dependencies. To address…

on-policy distillation (OPD)language-model agentsReAct interaction patternscross-domain effectsgeneral-reasoning supervisioncode-interpreter supervisionmathematical tool attemptsobservation acquisitionnegative transfersource-domain training trajectories
Upvotes
7
Authors
8
Published
Nov 14, 2026

Summary and abstract excerpt via Hugging Face Papers ↗ · arXiv.

Trending history

Agentic On-Policy Distillation Implicitly Reshapes LLMs' ReAct Patterns was first seen on Hugging Face · Daily papers on Nov 18, 2026. Since then it has appeared in 11 checks on 2 different days. Its best position was #18 of 30, reached on Nov 18, 21:28 UTC. It dropped off the list after Nov 19, 03:28 UTC.

Rank history

Position on Hugging Face · Daily papers at every check (lower is better).

#1#11#20#30 07:2812:2817:2822:2803:28 Nov 18, 07:28 UTC · #26Nov 18, 09:28 UTC · #21Nov 18, 11:28 UTC · #23Nov 18, 13:28 UTC · #22Nov 18, 15:28 UTC · #20Nov 18, 17:28 UTC · #20Nov 18, 19:28 UTC · #20Nov 18, 21:28 UTC · #18Nov 18, 23:28 UTC · #19Nov 19, 01:28 UTC · #20Nov 19, 03:28 UTC · #22

Last 14 days

Best rank per day (UTC).

Nov 9today

Metric history

Platform metric as reported by Hugging Face.

0369 07:2812:2817:2822:2803:28 Nov 18, 07:28 UTC · 6Nov 18, 09:28 UTC · 9Nov 18, 11:28 UTC · 9Nov 18, 13:28 UTC · 6Nov 18, 15:28 UTC · 6Nov 18, 17:28 UTC · 6Nov 18, 19:28 UTC · 6Nov 18, 21:28 UTC · 7Nov 18, 23:28 UTC · 7Nov 19, 01:28 UTC · 7Nov 19, 03:28 UTC · 7

Questions

Is Agentic On-Policy Distillation Implicitly Reshapes LLMs' ReAct Patterns trending on HF Daily Papers right now?

Not at the moment. It was last on the list on Nov 19, 03:28 UTC, after 2 days of trending.

What is the highest rank Agentic On-Policy Distillation Implicitly Reshapes LLMs' ReAct Patterns reached?

#18 on Hugging Face · Daily papers, on Nov 18, 21:28 UTC. Trendoday checks this list every 120 minutes.

How long has Agentic On-Policy Distillation Implicitly Reshapes LLMs' ReAct Patterns been trending?

It was first seen on Nov 18, 2026 and has been on the list on 2 days (11 checks) since.