{"slug": "distill-where-the-student-goes-teacher-regularized-rl-for-english-evidence-cross", "title": "Distill Where the Student Goes: Teacher-Regularized RL for English-Evidence Cross-Lingual RAG", "summary": "Researchers propose TR-RAG, a teacher-regularized reinforcement learning method for cross-lingual retrieval-augmented generation that addresses language drift and unreliable evidence use. The method combines reward optimization with on-policy distillation, improving language consistency and evidence-grounded correctness across three benchmarks. The teacher anchor prevents performance collapses in in-domain languages and improves evidence grounding in out-of-distribution languages.", "body_md": "arXiv:2607.02966v1 Announce Type: new\nAbstract: Cross-lingual retrieval-augmented generation (RAG) is often deployed in an English-evidence regime, where users query in diverse languages but retrieved passages remain English. In this setting, generation can fail despite strong base models: English evidence induces language drift (English or code-switching outputs) and models use evidence unreliably when producing non-English answers. We attribute these failures to two post-training challenges: (i) errors are prefix-dependent, so fixed-trajectory supervision suffers from prefix mismatch; and (ii) sequence-level (partly discrete / judge-based) rewards yield noisy credit assignment and high-variance updates. We propose TR-RAG, a teacher-regularized RL recipe that couples reward optimization with on-policy distillation on student-visited prefixes. A compact student samples on-policy answers, while a stronger frozen teacher is queried only on those prefixes and provides a prefix-wise student-to-teacher reverse-KL anchor. We further introduce a reward decomposition for English-evidence multilingual generation, combining language consistency, character 3-gram recall, and an LLM-judge score for evidence-grounded correctness. Across three benchmarks -- BioASQ-ENKB5, Hotpot-ENKB5, and naturally multilingual MKQA -- and two backbones, TR-RAG improves the composite of language adherence and evidence-grounded correctness over strong baselines. Crucially, the teacher anchor acts as a safety net: on in-domain languages it prevents the large language-consistency collapses (up to ~27 percentage points) that reward-only RL can suffer by drifting below even the base model, while on distant out-of-distribution languages -- where reward-only RL stalls at the base model's ceiling -- it still improves evidence grounding; and on character 3-gram recall the compact student sometimes surpasses its 70B teacher.", "url": "https://wpnews.pro/news/distill-where-the-student-goes-teacher-regularized-rl-for-english-evidence-cross", "canonical_source": "https://arxiv.org/abs/2607.02966", "published_at": "2026-07-07 04:00:00+00:00", "updated_at": "2026-07-07 04:01:59.586174+00:00", "lang": "en", "topics": ["large-language-models", "natural-language-processing", "ai-research", "machine-learning"], "entities": ["TR-RAG", "BioASQ-ENKB5", "Hotpot-ENKB5", "MKQA"], "alternates": {"html": "https://wpnews.pro/news/distill-where-the-student-goes-teacher-regularized-rl-for-english-evidence-cross", "markdown": "https://wpnews.pro/news/distill-where-the-student-goes-teacher-regularized-rl-for-english-evidence-cross.md", "text": "https://wpnews.pro/news/distill-where-the-student-goes-teacher-regularized-rl-for-english-evidence-cross.txt", "jsonld": "https://wpnews.pro/news/distill-where-the-student-goes-teacher-regularized-rl-for-english-evidence-cross.jsonld"}}