Learning from Trials and Errors: Reflective Test-Time Planning for Embodied LLMs
Yining Hong, Huang Huang, Manling Li, Li Fei-Fei, Jiajun Wu, Yejin Choi
https://arxiv.org/abs/2602.21198 https://arxiv.org/pdf/2602.21198 https://arxiv.org/html/2602.21198
arXiv:2602.21198v1 Announce Type: new
Abstract: Embodied LLMs endow robots with high-level task reasoning, but they cannot reflect on what went wrong or why, turning deployment into a sequence of independent trials where mistakes repeat rather than accumulate into experience. Drawing upon human reflective practitioners, we introduce Reflective Test-Time Planning, which integrates two modes of reflection: \textit{reflection-in-action}, where the agent uses test-time scaling to generate and score multiple candidate actions using internal reflections before execution; and \textit{reflection-on-action}, which uses test-time training to update both its internal reflection model and its action policy based on external reflections after execution. We also include retrospective reflection, allowing the agent to re-evaluate earlier decisions and perform model updates with hindsight for proper long-horizon credit assignment. Experiments on our newly-designed Long-Horizon Household benchmark and MuJoCo Cupboard Fitting benchmark show significant gains over baseline models, with ablative studies validating the complementary roles of reflection-in-action and reflection-on-action. Qualitative analyses, including real-robot trials, highlight behavioral correction through reflection.
toXiv_bot_toot
Study reveals hidden climate impact of digital industries #climate
🪤 France Bets on Carbon Capture as North Sea Rivals Surge Ahead
https://oilprice.com/Energy/Energy-General/France-Bets-on-Carbon-Capture-as-North-Sea-Rivals-Surge-Ahead.html
NFL proposes rule to assist possible replacement officials for 2026 season https://www.nytimes.com/athletic/7147157/2026/03/25/nfl-rule-proposal-replacement-officials/
The EU is looking for short fixes to reduce gas demand and has failed to find any.
https://www.euronews.com/my-europe/2026/03/23/eu-pushes-early-gas-refills-while-easing-storage-targets-on-iran-war
China is ex…
Ui, digitale Unabhängigkeit nimmt auch woanders Fahrt auf!
Leave big tech behind! How to replace Amazon, Google, X, Meta, Apple – and more
https://www.theguardian.com/technology/2026/feb/26/how-to-replac…
Leave big tech behind! How to replace Amazon, Google, X, Meta, Apple – and more;
A handful of companies monopolize the web, with unprecedented access to our data. But there are many more ethical – and often distinctively European – alternatives
https://www.theg…
NFL, referees union inch toward showdown as replacement officials loom
https://www.cbssports.com/nfl/news/nfl-referee-labor-dispute-replacement-officials-2026/
Tom Brady reveals he previously looked into potential NFL comeback https://www.nfl.com/news/tom-brady-reveals-he-previously-looked-into-potential-nfl-comeback
Lavonte David's remarkable run, plus questions for Maxx Crosby, Puka Nacua and replacement refs?! https://www.nytimes.com/athletic/7149599/2026/03/26/lavonte-davids-remarkable-run-plus-questions-f…