Tootfinder

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

A look at how tech executives use "delay, deny, and deflect" when pressed on user safety and what it means for journalism, as shame loses power over public life (Casey Newton/Platformer)
https://www.platformer.news/roblox-ceo-interview-backlash-analysis/<…

Roblox is a problem — but it’s a symptom of something worse
What is the role of tech journalism in a world where CEOs no longer feel shame?

Why We've Tried to Replace #Developers Every Decade Since 1969
https://www.caimito.net/en/blog/2025/12/07/the-recurring-dream-of-replacing-develop…

Ukraine repels massive Russian drone and missile attack: https://benborges.xyz/2025/12/23/ukraine-repels-massive-russian-drone.html

Replacing animals in science #AnimalRights

Replacing animals in science - Human Relevant Science
The UK Government has (11.11.2025) published its strategy for “Replacing animals in science”, to deliver on a key pledge from the Labour Party’s election manifesto to set out a clear plan to reduce and ultimately phase out the use of animals in research. The Alliance for Human Relevant Science welcomes this turning point in government-led […]

If you didn't get the Pip-Boy 3000 MkV, you can now get the Fallout 3/New Vegas edition of the Pip-Boy 3000.
▶️ Fallout Pip-Boy 3000 Replica – Official Bethesda Gear Store
https://gear.bethesda.net/products/fallout-pip-boy-3000-replica

Fallout Pip-Boy 3000 Replica

slashdot_threads: Slashdot thread replies (2007)
A network of replies among users of the website Slashdot on the various discussion threads on the site. Nodes are user accounts and each directed edge represents a reply from user i to user j. Replies are timestamped.
This network has 51083 nodes and 140778 edges.
Tags: Social, Communication, Unweighted, Multigraph, Timestamps

Zelensky reveals full 20-point peace plan draft backed by Ukraine, US: https://benborges.xyz/2025/12/24/zelensky-reveals-full-point-peace.html

OpenAI hires Arvind KC as chief people officer, replacing Julia Villagra who left in August 2025; KC has held senior roles at Roblox, Google, Palantir, and Meta (Agnee Ghosh/Bloomberg)
https://www.bloomberg.com/news/articles/202…

❗️20 points of peace: Zelenskyy reveals key political document of the war #shorts: https://benborges.xyz/2025/12/24/points-of-peace-zelenskyy-reveals.html