arXiv AI September 12, 2026 By Rui Zhu, Minglong Cao, Chenyu Zhou, Jianghao Lin, Dongdong Ge

Beyond Verified Answers: Solver-Informed Self-Distillation for Bootstrapping Operations Research Language Models

The paper introduces SOLID, a framework that enables operations research language models to self-improve without relying on verified answers or external evaluators. SOLID uses solver-generated artifacts from the model’s own rollouts to create pseudo-references, clustering objectives and applying group-relative advantages for dense self-supervision. Experiments on multiple OR benchmarks show that SOLID enhances solution accuracy for both general-purpose and OR-tuned models compared to outcome-only training.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv Machine Learning

2d ago

Negative Self-Distillation: Learning to Reason by Avoiding Flaws

Negative Self-Distillation (NSD) is a new framework for improving large language models by encouraging them to diverge from their own flawed reasoning rather than imitate privileged solutions. Unlike On-Policy Self-Distillation, which can suppress uncertainty and exploratory behavior, NSD generates a question‑specific negative condition (e.g., a careless reasoner) and uses a dynamic gating mechanism to target only reasoning‑critical tokens for penalization. This approach preserves foundational language capabilities while consistently outperforming OPSD and other label‑free self‑bootstrapping reinforcement learning baselines.

By Rongcan Pei, Zhepei Wei, Shuyao Xu, Xinyu Zhu, Wei-Lin Chen, Yu Meng

llms reinforcement-learning efficiency

arXiv AI

Jun 9

INFUSER: Influence-Guided Self-Evolution Improves Reasoning

arXiv:2606. 09052v1 Announce Type: cross Abstract: Self-evolution offers a scalable path to stronger reasoning: a pretrained language model improves itself with only minimal external supervision.

By Siyu Chen, Miao Lu, Beining Wu, Heejune Sheen, Fengzhuo Zhang, Shuangning Li, Zhiyuan Li, Jose Blanchet, Tianhao Wang, Zhuoran Yang

llms benchmarks

arXiv AI

Jul 1

From Search to Synthesis: Training LLMs as Zero-Shot Workflow Generators

arXiv:2606. 30704v1 Announce Type: cross Abstract: Large language models (LLMs) excel across a wide range of tasks, yet their instance-specific solutions often lack the structural consistency needed for reliable deployment.

By Gan Luo, Zihan Qin, Bin Dong, Wotao Yin

llms nlp reinforcement-learning fine-tuning benchmarks

arXiv AI

Jul 3

Neuron-Aware Data Selection for Annotation-Free LLM Self-Distillation

arXiv:2607. 02460v1 Announce Type: cross Abstract: Post-training large language models (LLMs) without real-world interaction feedback or human-labeled supervision remains challenging, particularly in specialized domains where expert annotations are costly to obtain.

By Zhuowei Chen, Xiang Lorraine Li

llms reinforcement-learning fine-tuning efficiency benchmarks

Hugging Face Trending Papers

3d ago

Negative Self-Distillation: Learning to Reason by Avoiding Flaws

Negative Self-Distillation (NSD) is a new framework for improving large language models by encouraging them to diverge from their own flawed reasoning rather than imitate privileged solutions. Unlike On-Policy Self-Distillation (OPSD), which can suppress uncertainty and exploratory behavior, NSD generates a question‑specific negative condition (e.g., a careless reasoner) and pushes the student’s distribution away from it. A dynamic gating mechanism isolates reasoning‑critical tokens so that only behavioral flaws are penalized, preserving linguistic capabilities, and empirical results show NSD consistently outperforms OPSD and other label‑free self‑bootstrapping RL baselines.

llms reinforcement-learning efficiency

arXiv Machine Learning

Jul 28

Self-Boosting Vision-Language Models with Noisy Student On-Policy Self-Distillation

arXiv:2607. 23125v1 Announce Type: new Abstract: Post-training enables vision-language models (VLMs) to understand human instructions and perform various downstream tasks.

By Shuai Wang, Daoan Zhang, Zhe Tang, Hao Cheng, Jiaheng Wei

llms reinforcement-learning efficiency multimodal benchmarks

One email a morning, machine-written

Related stories

Negative Self-Distillation: Learning to Reason by Avoiding Flaws

INFUSER: Influence-Guided Self-Evolution Improves Reasoning

From Search to Synthesis: Training LLMs as Zero-Shot Workflow Generators

Neuron-Aware Data Selection for Annotation-Free LLM Self-Distillation

Negative Self-Distillation: Learning to Reason by Avoiding Flaws

Self-Boosting Vision-Language Models with Noisy Student On-Policy Self-Distillation