Friday, August 7, 2026

1. RP-OPSD: Reasoning-Pivot-Guided On-Policy Self-Distillation for Multilingual Reasoning Transfer
   T2arXiv AI/CL/LG（论文）

   RP-OPSD proposes a self-distillation method for multilingual reasoning transfer. On-policy self-distillation (OPSD) provides dense token-level supervision but does not explicitly prioritize the reasoning signals that matter most for cross-lingual transfer. The method decomposes target-language reasoning into surface text and reasoning pivots, and uses the distributional shift between matched teacher views with and without an English reference solution as a proxy for pivots, enabling privileged distillation and reference anchoring. Experiments on mathematical reasoning benchmarks spanning 17 languages and multiple difficulty levels show that RP-OPSD outperforms strong multilingual reasoning baselines and OPSD variants. Analysis shows it concentrates privileged distillation on reasoning-control and problem-conditioned state-update tokens, downweighting surface realization tokens. Code is open-sourced.

—

—

—

1. Datasette 1.0a38 Fixes SQL Injection in Mixed Public/Private Table Setups
   T2Simon Willison 博客（Atom）

   Datasette 1.0a38 has been released, fixing a SQL injection vulnerability affecting instances that serve a mix of public and private tables in the same database. The bug allowed users with access to any public table to bypass the execute-sql permission restriction and gain read-only access to private tables via SQL injection. Administrators using this configuration are advised to disable execute-sql on that database. The fix is also available in Datasette 0.65.3. Such mixed deployments are likely rare, so the impact is limited.

2. Simon Willison on Technical Blogging: Lower Your Standards and Publish
   T2Simon Willison 博客（Atom）

   Simon Willison shares highlights from his interview with Cynthia Dunlop, with the core advice to 'lower your standards' and publish while still unhappy with the draft — otherwise you end up with a folder full of unpublished drafts. He discusses why he continues blogging, the most surprising impact, his proudest and hardest posts, and practical tips for beginners. The view challenges perfectionism and emphasizes the value of public sharing.