# ChemOPD: Multi-Teacher On-Policy Distillation for Multi-Task Chemical Reasoning

arXiv: https://arxiv.org/abs/2609.33838 · HTML: https://mlsanity.com/paper/2609.33838/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Prompting and chain-of-thought reasoning, one of 17 directions in LLM reasoning.
- Builds on 17 of its 39 references that are in the map.
- First posted on arXiv 27 Sep 2026.

## Placed in

- LLM reasoning / [Prompting and chain-of-thought reasoning](https://mlsanity.com/lr/prompting-and-chain-of-thought-reasoning/)
- Knowledge distillation / [On-policy and self-distillation for LLMs](https://mlsanity.com/distill/on-policy-and-self-distillation-for-llms/)

## Builds on (most cited first, 17 of 17)

- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Rethinking On-Policy Distillation of Large Language Models: Phenomenology, Mechanism, and Recipe](https://mlsanity.com/paper/2604.13016/) (2026)
- [A Survey on Knowledge Distillation of Large Language Models](https://mlsanity.com/paper/2402.13116/) (2024)
- [A Survey of On-Policy Distillation for Large Language Models](https://mlsanity.com/paper/2604.00626/) (2026)
- [LlaSMol: Advancing Large Language Models for Chemistry with a Large-Scale, Comprehensive, High-Quality Instruction Tuning Dataset](https://mlsanity.com/paper/2402.09391/) (2024)
- [ChemLLM: A Chemical Large Language Model](https://mlsanity.com/paper/2402.06852/) (2024)
- [TIP: Token Importance in On-Policy Distillation](https://mlsanity.com/paper/2604.14084/) (2026)
- [Adapting Auxiliary Losses Using Gradient Similarity](https://mlsanity.com/paper/1812.02224/) (2018)
- [MOPD: Multi-Teacher On-Policy Distillation for Capability Integration in LLM Post-Training](https://mlsanity.com/paper/2606.30406/) (2026)
- [ChemDFM-R: A Chemical Reasoning LLM Enhanced with Atomized Chemical Knowledge](https://mlsanity.com/paper/2507.21990/) (2025)
- [MAD-OPD: Breaking the Ceiling in On-Policy Distillation via Multi-Agent Debate](https://mlsanity.com/paper/2605.01347/) (2026)
- [Counteraction-Aware Multi-Teacher On-Policy Distillation for General Capability Recovery with Domain Preservation](https://mlsanity.com/paper/2605.27115/) (2026)
- [DanceOPD: On-Policy Generative Field Distillation](https://mlsanity.com/paper/2606.27377/) (2026)
- [Enhancing Chemical Reaction and Retrosynthesis Prediction with Large Language Model and Dual-task Learning](https://mlsanity.com/paper/2505.02639/) (2025)
- [Towards Principled Task Grouping for Multi-Task Learning](https://mlsanity.com/paper/2402.15328/) (2024)
- [MolEmb: Multimodal Large Language Models Can Be Strong Molecular Embedding Models](https://mlsanity.com/paper/2608.23646/) (2026)

## Abstract (arXiv)

Large language models are increasingly expected to support diverse chemical reasoning capabilities within a unified model. One approach is to develop specialized capabilities separately and consolidate them through multi-teacher on-policy distillation, but this raises two questions: how should specialization be organized, and how should specialist guidance be integrated? We introduce ChemOPD, which addresses both. We estimate task affinities from supervised fine-tuning gradients and solve a constrained mixed-integer program(MIP) to construct partially overlapping specialist groups. During distillation, we retain a generalist teacher trained on all tasks so that specialist guidance supplements rather than replaces its supervision. Our anchor-residual objective gradually increases the routed specialist's contribution on student-generated responses. On ChemCoTBench, affinity-guided specialization produces task-dependent gains over the generalist teacher and improves several capabilities beyond semantic task grouping. Yet stronger teacher-side performance does not automatically yield stronger students: with the same specialists and routes, anchor-residual OPD improves most reported metrics over specialist-only distillation and realizes a larger share of the available teacher gains. These results highlight specialization and capability integration as connected but distinct design problems in chemical reasoning.