Tootfinder

'Confirmation of Kremlin state policy' — Ukraine slams Russia's withdrawal from torture prevention convention: https://benborges.xyz/2025/09/29/confirmation-of-kremlin-state-policy.html

Trump Pivots Second Term Toward Foreign Policy (Wall Street Journal)
https://www.wsj.com/politics/policy/trump-pivots-second-term-toward-foreign-policy-08d66474
http://www.memeorandum.com/251030/p143#a251030p143

Random Policy Valuation is Enough for LLM Reasoning with Verifiable Rewards
Haoran He, Yuxiao Ye, Qingpeng Cai, Chen Hu, Binxing Jiao, Daxin Jiang, Ling Pan
https://arxiv.org/abs/2509.24981

Random Policy Valuation is Enough for LLM Reasoning with Verifiable Rewards
RL with Verifiable Rewards (RLVR) has emerged as a promising paradigm for improving the reasoning abilities of large language models (LLMs). Current methods rely primarily on policy optimization frameworks like PPO and GRPO, which follow generalized policy iteration that alternates between evaluating the current policy's value and improving the policy based on evaluation. While effective, they often suffer from training instability and diversity collapse, requiring complex heuristic tricks and …

🌍 The climate crisis demands urgent action. But which actions are best?
Decision makers face tough trade-offs:
Policy A lowers emissions at home but increases reliance on imports.
Policy B cuts emissions long-term but raises unemployment short-term.
Policy C boosts jobs now but increases emissions in the near term.
None of these choices are simple. A policy that looks good locally may increase global emissions, or its effects may depend on what other countries d…

A new policy document authored by seven Chinese government departments outlines plans to create a globally competitive brain-computer interface industry by 2030 (Emily Mullin/Wired)
https://www.wired.com/story/china-is-getting-serious-about-brain-compute…

China Is Building a Brain-Computer Interface Industry
A new policy document outlines China’s plan to create an internationally competitive BCI industry within five years, and proposes developing devices for both health and consumer uses.

One-Token Rollout: Guiding Supervised Fine-Tuning of LLMs with Policy Gradient
Rui Ming, Haoyuan Wu, Shoubo Hu, Zhuolun He, Bei Yu
https://arxiv.org/abs/2509.26313 https://

One-Token Rollout: Guiding Supervised Fine-Tuning of LLMs with Policy Gradient
Supervised fine-tuning (SFT) is the predominant method for adapting large language models (LLMs), yet it often struggles with generalization compared to reinforcement learning (RL). In this work, we posit that this performance disparity stems not just from the loss function, but from a more fundamental difference: SFT learns from a fixed, pre-collected dataset, whereas RL utilizes on-policy data sampled from the current policy. Building on this hypothesis, we introduce one-token rollout (OTR), …

🔗 Two guys hated using Comcast, so they built their own fiber ISP
https://arstechnica.com/tech-policy/2025/07/two-guys-hated-using-comcast-so-they-built-their-own-fiber-isp/

Two guys hated using Comcast, so they built their own fiber ISP
Brothers-in-law use construction knowledge to compete against Comcast in Michigan.

U-DiT Policy: U-shaped Diffusion Transformers for Robotic Manipulation
Linzhi Wu, Aoran Mei, Xiyue Wang, Guo-Niu Zhu, Zhongxue Gan
https://arxiv.org/abs/2509.24579 https://

U-DiT Policy: U-shaped Diffusion Transformers for Robotic Manipulation
Diffusion-based methods have been acknowledged as a powerful paradigm for end-to-end visuomotor control in robotics. Most existing approaches adopt a Diffusion Policy in U-Net architecture (DP-U), which, while effective, suffers from limited global context modeling and over-smoothing artifacts. To address these issues, we propose U-DiT Policy, a novel U-shaped Diffusion Transformer framework. U-DiT preserves the multi-scale feature fusion advantages of U-Net while integrating the global context…

This has Stephen Miller's fingerprints all over it...
"‘I’m not coming home’: Trump policy holds people in Ice custody without bail"
"Restaurant worker’s case shows how Trump administration is ‘inflicting the maximum punishment’, experts say"
https://www.t…

Incremental Policy Iteration for Unknown Nonlinear Systems with Stability and Performance Guarantees
Qingkai Meng, Fenglan Wang, Lin Zhao
https://arxiv.org/abs/2508.21367 https:…

Incremental Policy Iteration for Unknown Nonlinear Systems with Stability and Performance Guarantees
This paper proposes a general incremental policy iteration adaptive dynamic programming (ADP) algorithm for model-free robust optimal control of unknown nonlinear systems. The approach integrates recursive least squares estimation with linear ADP principles, which greatly simplifies the implementation while preserving adaptive learning capabilities. In particular, we develop a sufficient condition for selecting a discount factor such that it allows learning the optimal policy starting with an i…

Wow.
https://arstechnica.com/tech-policy/2025/10/man-finally-released-a-month-after-absurd-arrest-for-reposting-trump-meme/

Man finally released a month after absurd arrest for reposting Trump meme
Bodycam footage undermined sheriff’s “true threat” justification for the arrest.

Ok, now do circumcision:
Kennedy: No sufficient evidence for Tylenol-autism link
https://thehill.com/policy/healthcare/5582061-kennedy-trump-tylenol-autism/

Die Drohnenkrise (I) https://www.german-foreign-policy.com/news/detail/10135

Reusable Test Suites for Reinforcement Learning
J{\o}rn Eirik Betten, Quentin Mazouni, Dennis Gross, Pedro Lind, Helge Spieker
https://arxiv.org/abs/2508.21553 https://

Reusable Test Suites for Reinforcement Learning
Reinforcement learning (RL) agents show great promise in solving sequential decision-making tasks. However, validating the reliability and performance of the agent policies' behavior for deployment remains challenging. Most reinforcement learning policy testing methods produce test suites tailored to the agent policy being tested, and their relevance to other policies is unclear. This work presents Multi-Policy Test Case Selection (MPTCS), a novel automated test suite selection method for RL en…

Economic Competition, EU Regulation, and Executive Orders: A Framework for Discussing AI Policy Implications in CS Courses
James Weichert, Hoda Eldardiry
https://arxiv.org/abs/2509.25524

Economic Competition, EU Regulation, and Executive Orders: A Framework for Discussing AI Policy Implications in CS Courses
The growth and permeation of artificial intelligence (AI) technologies across society has drawn focus to the ways in which the responsible use of these technologies can be facilitated through AI governance. Increasingly, large companies and governments alike have begun to articulate and, in some cases, enforce governance preferences through AI policy. Yet existing literature documents an unwieldy heterogeneity in ethical principles for AI governance, while our own prior research finds that disc…

PRPO: Paragraph-level Policy Optimization for Vision-Language Deepfake Detection
Tuan Nguyen, Naseem Khan, Khang Tran, NhatHai Phan, Issa Khalil
https://arxiv.org/abs/2509.26272

PRPO: Paragraph-level Policy Optimization for Vision-Language Deepfake Detection
The rapid rise of synthetic media has made deepfake detection a critical challenge for online safety and trust. Progress remains constrained by the scarcity of large, high-quality datasets. Although multimodal large language models (LLMs) exhibit strong reasoning capabilities, their performance on deepfake detection is poor, often producing explanations that are misaligned with visual evidence or hallucinatory. To address this limitation, we introduce a reasoning-annotated dataset for deepfake …

Multi-Agent Guided Policy Search for Non-Cooperative Dynamic Games
Jingqi Li, Gechen Qu, Jason J. Choi, Somayeh Sojoudi, Claire Tomlin
https://arxiv.org/abs/2509.24226 https://

Multi-Agent Guided Policy Search for Non-Cooperative Dynamic Games
Multi-agent reinforcement learning (MARL) optimizes strategic interactions in non-cooperative dynamic games, where agents have misaligned objectives. However, data-driven methods such as multi-agent policy gradients (MA-PG) often suffer from instability and limit-cycle behaviors. Prior stabilization techniques typically rely on entropy-based exploration, which slows learning and increases variance. We propose a model-based approach that incorporates approximate priors into the reward function a…

Judges really don't like this detention policy (Politico)
https://www.politico.com/newsletters/west-wing-playbook-remaking-government/2025/10/31/judges-really-dont-like-this-detention-policy-00631871
http://www.memeorandum.com/251031/p134#a251031p134

This brings to mind Gladstone's quote about judging a society based on how it treats its dead. We've got to do better. #JusticeReform #PrisonReform

These Families Wanted to Lay Their Loved Ones to Rest. They Had to Bring Them Home From Prison First.
Policies that dictate what happens after a death behind bars often add new layers of pain — and financial and logistical challenges — for those left behind.

Security Friction Quotient for Zero Trust Identity Policy with Empirical Validation
Michel Youssef
https://arxiv.org/abs/2509.22663 https://arxiv.org/pdf/2…

Security Friction Quotient for Zero Trust Identity Policy with Empirical Validation
We define a practical method to quantify the trade-off between security and operational friction in modern identity-centric programs. We introduce the Security Friction Quotient (SFQ), a bounded composite index that combines a residual-risk estimator with empirically grounded friction terms (latency, failure rate, and helpdesk impact). We establish clarity properties (boundedness, monotonic response, and weight identifiability) with short proofs, then evaluate widely used Conditional Access pol…

RoboInspector: Unveiling the Unreliability of Policy Code for LLM-enabled Robotic Manipulation
Chenduo Ying, Linkang Du, Peng Cheng, Yuanchao Shu
https://arxiv.org/abs/2508.21378

RoboInspector: Unveiling the Unreliability of Policy Code for LLM-enabled Robotic Manipulation
Large language models (LLMs) demonstrate remarkable capabilities in reasoning and code generation, enabling robotic manipulation to be initiated with just a single instruction. The LLM carries out various tasks by generating policy code required to control the robot. Despite advances in LLMs, achieving reliable policy code generation remains a significant challenge due to the diverse requirements of real-world tasks and the inherent complexity of user instructions. In practice, different users …

Vegetables. Monetary policy. Facebook: How one post about a central bank decision fired up Ukraine's top economists: https://benborges.xyz/2025/10/29/vegetables-monetary-policy-facebook-how.html

I have not ordered PCBs in a while...
"U.S.Tariff Policy Update: Products exported from China to the U.S. will now be levied a duty rate of 55% to 112.5%."
Fuuuuuuuuuuk...

Odyssey: Adaptive Policy Selection for Resilient Distributed Training
Yuhang Zhou, Zhibin Wang, Peng Jiang, Haoran Xia, Junhe Lu, Qianyu Jiang, Rong Gu, Hengxi Xu, Xinjing Huang, Guanghuan Fang, Zhiheng Hu, Jingyi Zhang, Yongjin Cai, Jian He, Chen Tian
https://arxiv.org/abs/2508.21613

Odyssey: Adaptive Policy Selection for Resilient Distributed Training
Training large language models faces frequent interruptions due to various faults, demanding robust fault-tolerance. Existing backup-free methods, such as redundant computation, dynamic parallelism, and data rerouting, each incur performance penalties, whether from ongoing overhead, lengthy reconfigurations, or post-recovery inefficiencies. We propose Odyssey, an adaptive fault-tolerant system that intelligently selects optimal recovery strategies when a failure occurs. Odyssey achieves this th…

What Makes LLM Agent Simulations Useful for Policy? Insights From an Iterative Design Engagement in Emergency Preparedness
Yuxuan Li, Sauvik Das, Hirokazu Shirado
https://arxiv.org/abs/2509.21868

What Makes LLM Agent Simulations Useful for Policy? Insights From an Iterative Design Engagement in Emergency Preparedness
There is growing interest in using Large Language Models as agents (LLM agents) for social simulations to inform policy, yet real-world adoption remains limited. This paper addresses the question: How can LLM agent simulations be made genuinely useful for policy? We report on a year-long iterative design engagement with a university emergency preparedness team. Across multiple design iterations, we iteratively developed a system of 13,000 LLM agents that simulate crowd movement and communicatio…

Harmeet Dhillon, a former Trump lawyer placed in charge of the Justice Department’s Civil Rights Division,
bragged to the Wall Street Journal that she “wakes up around 6am and begins her workday scrolling through X, searching for claims of discrimination …
After spotting ‘a list of new horrors,’ she said,
‘I text my deputies, and we assign cases, and we get cranking.’”
Dhillon’s office has been bleeding staff, since experienced lawyers are largely disinclined to waste…

The triumph of bullshit
Internet conspiracy theories are now official government policy.

LinkedIn, plattformen för professionellt nätverkande, uppdaterade för några dagar sedan sin policy gällande hatiskt och kränkande innehåll. Specifikt har man tagit bort formuleringar som inneburit ett tydligt ställningstagande mot trakasserier riktade mot transpersoner.

Läs hela inlägget: https://axbom.se/linkedin-raderar-tydligt-stod-for-transpersoner/

ORPO-Distill: Mixed-Policy Preference Optimization for Cross-Architecture LLM Distillation
Aasheesh Singh, Vishal Vaddina, Dagnachew Birru
https://arxiv.org/abs/2509.25100 https…

ORPO-Distill: Mixed-Policy Preference Optimization for Cross-Architecture LLM Distillation
We introduce ORPO-Distill, a general-purpose method for cross-architecture LLM distillation that formulates the problem as a preference optimization task. Un- like standard CoT distillation, the approach transfers knowledge through diverse reasoning traces. It employs an Odds-Ratio Preference Optimization objective that contrasts teacher and student traces for more effective learning, and adopts a mixed-policy strategy for utilizing student-generated outputs, outperforming both off- and on-poli…

Apple übernimmt Entwickler des Open Policy Agents
Die Erfinder des Open Policy Agents wechseln nach Cupertino: Apple kauft damit Expertise in der Open-Source-Software, die unter Kontrolle der CNCF bleibt.
https://www.

Improving Aviation Safety Analysis: Automated HFACS Classification Using Reinforcement Learning with Group Relative Policy Optimization
Arash Ahmadi, Sarah Sharif, Yaser Banad
https://arxiv.org/abs/2508.21201

Improving Aviation Safety Analysis: Automated HFACS Classification Using Reinforcement Learning with Group Relative Policy Optimization
Analyzing the human factors behind aviation accidents is crucial for preventing future incidents, yet traditional methods using the Human Factors Analysis and Classification System (HFACS) are limited by scalability and consistency. To address this, we introduce an automated HFACS classification framework for aviation safety analysis that utilizes Reinforcement Learning with Group Relative Policy Optimization (GRPO) to fine-tune a Llama-3.1 8B language model. Our approach incorporates a multi-c…

Anita Bryant would absolutely line up behind Trump, and not because of policy nuance, but because they’re united by the same obsession: pushing LGBTQIA people out of public life.
Bryant built an entire career on vilifying queer people in the 1970s, weaponizing her orange juice smile as a mask for hate. Trump, decades later, recycled the same homophobia and transphobia at a national scale, wrapping it up in authoritarianism and reactionary politics. They’re cut from the same cloth, one…

#Anthropic plant, künftig neue #Chatverläufe und #Codingsitzungen zur Verbesserung seiner KI-Modelle zu nutzen-
Nutzer müssen nicht aktiv widersprechen. Die

Anthropic will start training its AI models on chat transcripts
Anthropic will start training its AI models on user data, including chat transcripts and coding sessions, and extending its data retention policy to five years for active users — unless users choose to opt out.

Pope Leo XIV: Support for death penalty is ‘not really pro-life’
https://thehill.com/policy/international/5530335-pope-leo-abortion-death-penalty-pro-life/

ICE has policies??? That IS news!😏
Revealed: ICE violates its own policy by holding people in secretive rooms for days or weeks
https://www.theguardian.com/us-news/2025/oct/30/ice-hidden-detention-sites

Die Drohnenkrise (II) https://www.german-foreign-policy.com/news/detail/10137

Conservatives love to claim #liberal advocacy for immigration is a radical leftist policy.

Immigration has always been an American value: #Conservatives are the ones radicalized & don’t even realize it.
Watch

Seems like a good day to revisit this post about privacy-preserving age verification. (tl;dr -> you can't have it) https://www.techdirt.com/2025/08/19/privacy‑preserving-age-verification-falls-apart-on-contact-with…

Privacy‑Preserving Age Verification Falls Apart On Contact With Reality
Here we go again. Whenever policy makers insist that there’s some “nerd harder” solution to tricky societal problems, actual experts have to spend a ridiculous amount of time explaining basic reali…

Yes, of course we’re Agile. We have Jira!
Oh no sorry, sprints are enforced at 2 week length per corporate policy.
What’s that? No sorry, teams can’t modify task fields, you’d need to ask a Jira admin, but they don’t like to change things.
Swimlanes? Oh dear no, that’s the Jira admin as well. You see it’s important we have a strictly uniform implementation of our Agile workflow across the org.
(And they don’t even hear the irony themselves…)

Spotify updates its AI policy, including adopting the upcoming DDEX standard to label and identify AI music and rolling out a new music spam filter (Sarah Perez/TechCrunch)
https://techcrunch.com/2025/09/25/spotify-updates-ai-policy-to-label-trac…

Spotify to label AI music, filter spam and more in AI policy change | TechCrunch
Spotify is launching a music spam filter, labeling AI tracks, and clarifying its rules around AI voice clones.

Policy Optimization in Robust Control: Weak Convexity and Subgradient Methods
Yuto Watanabe, Feng-Yi Liao, Yang Zheng
https://arxiv.org/abs/2509.25633 https://

Policy Optimization in Robust Control: Weak Convexity and Subgradient Methods
Robust control seeks stabilizing policies that perform reliably under adversarial disturbances, with $\mathcal{H}_\infty$ control as a classical formulation. It is known that policy optimization of robust $\mathcal{H}_\infty$ control naturally lead to nonsmooth and nonconvex problems. This paper builds on recent advances in nonsmooth optimization to analyze discrete-time static output-feedback $\mathcal{H}_\infty$ control. We show that the $\mathcal{H}_\infty$ cost is weakly convex over any con…

Sci2Pol: Evaluating and Fine-tuning LLMs on Scientific-to-Policy Brief Generation
Weimin Wu, Alexander C. Furnas, Eddie Yang, Gefei Liu, Akhil Pandey Akella, Xuefeng Song, Dashun Wang, Han Liu
https://arxiv.org/abs/2509.21493

Sci2Pol: Evaluating and Fine-tuning LLMs on Scientific-to-Policy Brief Generation
We propose Sci2Pol-Bench and Sci2Pol-Corpus, the first benchmark and training dataset for evaluating and fine-tuning large language models (LLMs) on policy brief generation from a scientific paper. We build Sci2Pol-Bench on a five-stage taxonomy to mirror the human writing process: (i) Autocompletion, (ii) Understanding, (iii) Summarization, (iv) Generation, and (v) Verification. It features 18 tasks in multiple-choice and open-ended formats. Specifically, for the Generation stage, we show that…

Atlantis-horizon.eu
3 policy briefs
August 2025
site: https://www.atlantis-horizon.eu/building-europes-resilience-3-atlantis-policy-briefs/

Sampling Complexity of TD and PPO in RKHS
Lu Zou, Wendi Ren, Weizhong Zhang, Liang Ding, Shuang Li
https://arxiv.org/abs/2509.24991 https://arxiv.org/pdf/2…

Sampling Complexity of TD and PPO in RKHS
We revisit Proximal Policy Optimization (PPO) from a function-space perspective. Our analysis decouples policy evaluation and improvement in a reproducing kernel Hilbert space (RKHS): (i) A kernelized temporal-difference (TD) critic performs efficient RKHS-gradient updates using only one-step state-action transition samples; (ii) a KL-regularized, natural-gradient policy step exponentiates the evaluated action-value, recovering a PPO/TRPO-style proximal update in continuous state-action spaces.…

NFL fines Ravens $100K for incorrectly listing Lamar Jackson on Week 8 injury report https://www.nfl.com/news/nfl-fines-ravens-100k-incorrectly-listing-lamar-jackson-week-8-injury-report

At least I got the day job to implement a "no AI browsers on work machines" policy.
If you CISO has not released a policy forbidding AI browsers at your company they are asleep at the wheel.
https://mstdn.social/@Hackread/115448122449051028

Doctors at the US Department of Veterans Affairs to be barred from performing #abortions, even in cases of #rape and incest, under new rules proposed by the Trump administration. New #Drumpf policy edict to affect

Institutional Policy Pathways for Supporting Research Software: Global Trends and Local Practices
Michelle Barker, Jeremy Cohen, Pedro Hern\'andez Serrano, Daniel S. Katz, Kim Martin, Dan Rudmann, Hugh Shanahan
https://arxiv.org/abs/2509.26422

Institutional Policy Pathways for Supporting Research Software: Global Trends and Local Practices
As research software becomes increasingly central to modern science, research-performing organisations (RPOs) need to ensure that their investment in people, skills and infrastructure around research software produces sustainable and maintainable software that improves the research they perform, which in turn improves the overall institution and its reputation and funding, for example, by competing with peers who lack this approach. However, research institution management and recognition of re…

Replaced article(s) found for cs.AI. https://arxiv.org/list/cs.AI/new
[1/4]:
- Policy Expansion for Bridging Offline-to-Online Reinforcement Learning
Haichao Zhang, We Xu, Haonan Yu

GSPR: Aligning LLM Safeguards as Generalizable Safety Policy Reasoners
Haoran Li, Yulin Chen, Jingru Zeng, Hao Peng, Huihao Jing, Wenbin Hu, Xi Yang, Ziqian Zeng, Sirui Han, Yangqiu Song
https://arxiv.org/abs/2509.24418

GSPR: Aligning LLM Safeguards as Generalizable Safety Policy Reasoners
As large language models (LLMs) are increasingly integrated into numerous applications across various domains, LLMs' safety becomes a critical concern for both application developers and intended users. Currently, great efforts have been made to develop safety benchmarks with fine-grained taxonomies. However, these benchmarks' taxonomies are disparate with different safety policies. Thus, existing safeguards trained on these benchmarks are either coarse-grained to only distinguish between safe …

ReSeFlow: Rectifying SE(3)-Equivariant Policy Learning Flows
Zhitao Wang, Yanke Wang, Jiangtao Wen, Roberto Horowitz, Yuxing Han
https://arxiv.org/abs/2509.22695 https://…

ReSeFlow: Rectifying SE(3)-Equivariant Policy Learning Flows
Robotic manipulation in unstructured environments requires the generation of robust and long-horizon trajectory-level policy with conditions of perceptual observations and benefits from the advantages of SE(3)-equivariant diffusion models that are data-efficient. However, these models suffer from the inference time costs. Inspired by the inference efficiency of rectified flows, we introduce the rectification to the SE(3)-diffusion models and propose the ReSeFlow, i.e., Rectifying SE(3)-Equivari…

3-star Lt. Gen. Joe McGee steps down amid reported friction with Pete Hegseth, Dan Caine
https://thehill.com/policy/defense/5581245-three-star-general-mcghee-retires/

Entscheidung erneut verschoben https://www.german-foreign-policy.com/news/detail/10095

DOGE staffer set to replace admiral leading Navy's research office (Ellen Mitchell/The Hill)
https://thehill.com/policy/defense/5581695-navy-chief-research-replaced/
http://www.memeorandum.com/251030/p126#a251030p126

Developer Insights into Designing AI-Based Computer Perception Tools
Maya Guhan (Center for Ethics and Health Policy, Baylor College of Medicine, Houston, TX, USA), Meghan E. Hurley (Center for Ethics and Health Policy, Baylor College of Medicine, Houston, TX, USA), Eric A. Storch (Department of Psychiatry and Behavioral Sciences, Baylor College of Medicine, Houston, TX, USA), John Herrington (Department of Child and Adolescent Psychiatry and Behavioral Sciences, Children's Hospita…

Developer Insights into Designing AI-Based Computer Perception Tools
Artificial intelligence (AI)-based computer perception (CP) technologies use mobile sensors to collect behavioral and physiological data for clinical decision-making. These tools can reshape how clinical knowledge is generated and interpreted. However, effective integration of these tools into clinical workflows depends on how developers balance clinical utility with user acceptability and trustworthiness. Our study presents findings from 20 in-depth interviews with developers of AI-based CP to…

SPARK: Synergistic Policy And Reward Co-Evolving Framework
Ziyu Liu, Yuhang Zang, Shengyuan Ding, Yuhang Cao, Xiaoyi Dong, Haodong Duan, Dahua Lin, Jiaqi Wang
https://arxiv.org/abs/2509.22624

SPARK: Synergistic Policy And Reward Co-Evolving Framework
Recent Large Language Models (LLMs) and Large Vision-Language Models (LVLMs) increasingly use Reinforcement Learning (RL) for post-pretraining, such as RL with Verifiable Rewards (RLVR) for objective tasks and RL from Human Feedback (RLHF) for subjective tasks. However, RLHF incurs high costs and potential reward-policy mismatch due to reliance on human preferences, while RLVR still wastes supervision by discarding rollouts and correctness signals after each update. To address these challenges,…

Trump’s 50 percent tariffs landed like a declaration of economic war on India,
undercutting enormous investments made by American companies to hedge their dependency on China.
India’s hard work to present itself to the world as the best alternative to Chinese factories
— what business executives and big money financiers have embraced as part of the China Plus One strategy
— has been left in tatters.
Now, less than a week since the tariffs took full effect, offic…

India Was the Economic Alternative to China. Trump Ended That.
A lurch in policy has shaken the India-U.S. economic alliance against China, leaving India little choice but to consider reversing its own strategy.

Spotify updates its AI policy, including adopting the upcoming DDEX standard to label and identify AI music and rolling out a new music spam filter (Sarah Perez/TechCrunch)
https://techcrunch.com/2025/09/25/spotify-updates-ai-policy-to-label-trac…

Spotify to label AI music, filter spam and more in AI policy change | TechCrunch
Spotify is launching a music spam filter, labeling AI tracks, and clarifying its rules around AI voice clones.

Ukraine developing 'careful' policy on sharing 'priceless' wartime data, minister tells Reuters: https://benborges.xyz/2025/08/27/ukraine-developing-careful-policy-on.html

Transatlantische Weltraumrivalen https://www.german-foreign-policy.com/news/detail/10179

Lankford on autism and vaccines: 'I don't connect those two' (Tara Suter/The Hill)
https://thehill.com/policy/healthcare/5479434-lankford-on-autism-and-vaccines-i-dont-connect-those-two/
http://www.memeorandum.com/250831/p78#a250831p78

Detecting YouTube Scam Videos via Multimodal Signals and Policy Reasoning
Ummay Kulsum, Aafaq Sabir, Abhinaya S. B., Anupam Das
https://arxiv.org/abs/2509.23418 https://

Detecting YouTube Scam Videos via Multimodal Signals and Policy Reasoning
YouTube has emerged as a dominant platform for both information dissemination and entertainment. However, its vast accessibility has also made it a target for scammers, who frequently upload deceptive or malicious content. Prior research has documented a range of scam types, and detection approaches rely primarily on textual or statistical metadata. Although effective to some extent, these signals are easy to evade and potentially overlook other modalities, such as visual cues. In this study,…

Unlocking the Potential of Soft Actor-Critic for Imitation Learning
Nayari Marie Lessa, Melya Boukheddimi, Frank Kirchner
https://arxiv.org/abs/2509.24539 https://

Unlocking the Potential of Soft Actor-Critic for Imitation Learning
Learning-based methods have enabled robots to acquire bio-inspired movements with increasing levels of naturalness and adaptability. Among these, Imitation Learning (IL) has proven effective in transferring complex motion patterns from animals to robotic systems. However, current state-of-the-art frameworks predominantly rely on Proximal Policy Optimization (PPO), an on-policy algorithm that prioritizes stability over sample efficiency and policy generalization. This paper proposes a novel IL f…

Beyond expected value: geometric mean optimization for long-term policy performance in reinforcement learning
Xinyi Sheng, Dominik Baumann
https://arxiv.org/abs/2508.21443 https…

Beyond expected value: geometric mean optimization for long-term policy performance in reinforcement learning
Reinforcement learning (RL) algorithms typically optimize the expected cumulative reward, i.e., the expected value of the sum of scalar rewards an agent receives over the course of a trajectory. The expected value averages the performance over an infinite number of trajectories. However, when deploying the agent in the real world, this ensemble average may be uninformative for the performance of individual trajectories. Thus, in many applications, optimizing the long-term performance of individ…

Trump won't send 'high level' representatives to COP30 climate summit (Rachel Frazin/The Hill)
https://thehill.com/policy/energy-environment/5583954-trump-cop30-brazil-united-nations-climate-summit/
http://www.memeorandum.com/251031/p129#a251031p129

Ravens fined $100K for Jackson injury 'mistake' https://www.espn.com/nfl/story/_/id/46794185/ravens-fined-100k-injury-report-violation-lamar-jackson

The Pentagon's openness to the press was never a favor to journalists, but an obligation to the country, and its new restrictions make its leaders appear weak (Nancy A. Youssef/The Atlantic)
https://www.theatlant…

Why Is the Pentagon Afraid of the Press?
Pete Hegseth’s department is imposing restrictions that threaten to limit media scrutiny.

Former CDC official on RFK Jr. policies: 'I only see harm coming' (Rachel Scully/The Hill)
https://thehill.com/policy/healthcare/5479421-former-cdc-official-kennedy-alarm/
http://www.memeorandum.com/250831/p77#a250831p77

Breaks the law: "the policy is aimed primarily at organizations that work with immigrants and transgender youth"
Trump administration completes new rules on public service loan forgiveness | AP News
https://apnews.com/article/public-service-loan-forgiveness-debt-cancellation-62536bdb25f04b5138d2d0a2aeec4e58

Iterative Inference in a Chess-Playing Neural Network
Elias Sandmann, Sebastian Lapuschkin, Wojciech Samek
https://arxiv.org/abs/2508.21380 https://arxiv.o…

Iterative Inference in a Chess-Playing Neural Network
Do neural networks build their representations through smooth, gradual refinement, or via more complex computational processes? We investigate this by extending the logit lens to analyze the policy network of Leela Chess Zero, a superhuman chess engine. We find strong monotonic trends in playing strength and puzzle-solving ability across layers, yet policy distributions frequently follow non-smooth trajectories. Evidence for this includes correct puzzle solutions that are discovered early but s…

Sanders pressures RFK Jr. to resign as Health chief after CDC chaos (Filip Timotija/The Hill)
https://thehill.com/policy/healthcare/5479097-sanders-pressures-rfk-jk-resign-health-chief-cdc-chaos/
http://www.memeorandum.com/250831/p15#a250831p15

Group Critical-token Policy Optimization for Autoregressive Image Generation
Guohui Zhang, Hu Yu, Xiaoxiao Ma, JingHao Zhang, Yaning Pan, Mingde Yao, Jie Xiao, Linjiang Huang, Feng Zhao
https://arxiv.org/abs/2509.22485

Group Critical-token Policy Optimization for Autoregressive Image Generation
Recent studies have extended Reinforcement Learning with Verifiable Rewards (RLVR) to autoregressive (AR) visual generation and achieved promising progress. However, existing methods typically apply uniform optimization across all image tokens, while the varying contributions of different image tokens for RLVR's training remain unexplored. In fact, the key obstacle lies in how to identify more critical image tokens during AR generation and implement effective token-wise optimization for them. T…

Learning Agile Gate Traversal via Analytical Optimal Policy Gradient
Tianchen Sun, Bingheng Wang, Longbin Tang, Yichao Gao, Lin Zhao
https://arxiv.org/abs/2508.21592 https://

Learning Agile Gate Traversal via Analytical Optimal Policy Gradient
Traversing narrow gates presents a significant challenge and has become a standard benchmark for evaluating agile and precise quadrotor flight. Traditional modularized autonomous flight stacks require extensive design and parameter tuning, while end-to-end reinforcement learning (RL) methods often suffer from low sample efficiency and limited interpretability. In this work, we present a novel hybrid framework that adaptively fine-tunes model predictive control (MPC) parameters online using outp…

“Opening locks” might not sound like scintillating social media content,
but Trevor McNally has turned lock-busting into online gold.
A former US Marine Staff Sergeant, McNally today has more than 7 million followers and has amassed more than 2 billion views
just by showing how easy it is to open many common locks by slapping, picking, or shimming them

10M people watched a YouTuber shim a lock; the lock company sued him. Bad idea.
It’s still legal to pick locks, even when you swing your legs.

Donald Trump is a globalist (Joshua Keating/Vox)
https://www.vox.com/world-politics/459513/trump-globalist-foreign-policy
http://www.memeorandum.com/250901/p28#a250901p28

New Jersey GOP Nominee Opposes Citizenship Policy That Made His Family American - National Memo
https://www.nationalmemo.com/birthright-citizenship-2674255850

White House Withdraws Nominee to Lead Bureau of Labor Statistics (Wall Street Journal)
https://www.wsj.com/politics/policy/white-house-withdraws-nominee-to-lead-bureau-of-labor-statistics-ceeae242?st=gxmRjy&reflink=desktopwebshare_permalink
http://www.memeorandum.com/250930/p172#a250930p172

MSG: Multi-Stream Generative Policies for Sample-Efficient Robotic Manipulation
Jan Ole von Hartz, Lukas Schweizer, Joschka Boedecker, Abhinav Valada
https://arxiv.org/abs/2509.24956

MSG: Multi-Stream Generative Policies for Sample-Efficient Robotic Manipulation
Generative robot policies such as Flow Matching offer flexible, multi-modal policy learning but are sample-inefficient. Although object-centric policies improve sample efficiency, it does not resolve this limitation. In this work, we propose Multi-Stream Generative Policy (MSG), an inference-time composition framework that trains multiple object-centric policies and combines them at inference to improve generalization and sample efficiency. MSG is model-agnostic and inference-only, hence widely…

„Buy European“ https://www.german-foreign-policy.com/news/detail/10133

EPO: Entropy-regularized Policy Optimization for LLM Agents Reinforcement Learning
Xu Wujiang, Wentian Zhao, Zhenting Wang, Li Yu-Jhe, Jin Can, Jin Mingyu, Mei Kai, Wan Kun, Metaxas Dimitris
https://arxiv.org/abs/2509.22576

EPO: Entropy-regularized Policy Optimization for LLM Agents Reinforcement Learning
Training LLM agents in multi-turn environments with sparse rewards, where completing a single task requires 30+ turns of interaction within an episode, presents a fundamental challenge for reinforcement learning. We identify a critical failure mode unique to this setting: the exploration-exploitation cascade failure. This cascade begins with early-stage policy premature convergence, where sparse feedback causes agents to commit to flawed, low-entropy strategies. Subsequently, agents enter late-…

Trump's Vow to Be 'President of Peace' Falls Short in Ukraine and Gaza (Wall Street Journal)
https://www.wsj.com/politics/policy/trumps-vow-to-be-president-of-peace-falls-short-in-ukraine-and-gaza-27b8731e
http://www.memeorandum.com/250801/p3#a250801p3

The only thing that drives every single Republican policy - Alternet.org
https://archive.ph/I7rO6

"the Fedora Council has approved the latest version of the AI-Assisted Contributions policy formally".
Whoever contributes the code is required to be fully transparent on what AI tool has been used for it.
https://www.gamingonlinux.co…

Fedora Linux project agrees to allow AI-assisted contributions with a new policy
AI generation, it's everywhere! And now it's going to be formally accepted into Fedora Linux, as per the latest approved change with a new policy.

Taxes, Psychedelics, Crypto: Kyrsten Sinema Sees Opportunity Under Trump (Wall Street Journal)
https://www.wsj.com/politics/policy/kyrsten-sinema-trump-administration-0bad2b72?st=zYJsN1&reflink=desktopwebshare_permalink
http://www.memeorandum.com/250831/p115#a250831p115

Kriegstüchtige Geheimdienste https://www.german-foreign-policy.com/news/detail/10177

From Parameters to Behavior: Unsupervised Compression of the Policy Space
Davide Tenedini, Riccardo Zamboni, Mirco Mutti, Marcello Restelli
https://arxiv.org/abs/2509.22566 http…

From Parameters to Behavior: Unsupervised Compression of the Policy Space
Despite its recent successes, Deep Reinforcement Learning (DRL) is notoriously sample-inefficient. We argue that this inefficiency stems from the standard practice of optimizing policies directly in the high-dimensional and highly redundant parameter space $Θ$. This challenge is greatly compounded in multi-task settings. In this work, we develop a novel, unsupervised approach that compresses the policy parameter space $Θ$ into a low-dimensional latent space $\mathcal{Z}$. We train a generativ…

Trump casts doubt on Putin-Zelensky meeting: 'Maybe they have to fight a little longer' (Filip Timotija/The Hill)
https://thehill.com/policy/international/5478995-donald-trump-vladimir-putin-volodymyr-zelensky-meeting-prospects/
http://www.memeorandum.com/250831/p7#a250831p7

Trump Triples Down on Lawfare (Kimberley A. Strassel/Wall Street Journal)
https://www.wsj.com/opinion/trump-triples-down-on-lawfare-politics-policy-democrats-opponents-79311d06
http://www.memeorandum.com/250831/p13#a250831p13

"The calls are part of a broader strategy by the Trump administration to ... turn the state department into an anti-immigration machine. "In the Trump administration migration issues have become a key pillar of our foreign policy, and this may be here to stay for the rest of the century,” a senior state department official told The Guardian."
**Stephen Miller holding secret calls with diplomats every day to push Trump’s immigration agenda across the globe**
https://www.independent.co.uk/news/world/americas/us-politics/stephen-miller-immigration-diplomats-secret-calls-b2856464.html

Learning on the Fly: Rapid Policy Adaptation via Differentiable Simulation
Jiahe Pan, Jiaxu Xing, Rudolf Reiter, Yifan Zhai, Elie Aljalbout, Davide Scaramuzza
https://arxiv.org/abs/2508.21065

Learning on the Fly: Rapid Policy Adaptation via Differentiable Simulation
Learning control policies in simulation enables rapid, safe, and cost-effective development of advanced robotic capabilities. However, transferring these policies to the real world remains difficult due to the sim-to-real gap, where unmodeled dynamics and environmental disturbances can degrade policy performance. Existing approaches, such as domain randomization and Real2Sim2Real pipelines, can improve policy robustness, but either struggle under out-of-distribution conditions or require costly…

„Wunsch nach klarer Friedenspolitik“ https://www.german-foreign-policy.com/news/detail/10176

Inferring Effects of Major Events through Discontinuity Forecasting of Population Anxiety
Siddharth Mangalik, Ojas Deshpande, Adithya V. Ganesan, Sean A. P. Clouston, H. Andrew Schwartz
https://arxiv.org/abs/2508.21722

Inferring Effects of Major Events through Discontinuity Forecasting of Population Anxiety
Estimating community-specific mental health effects of local events is vital for public health policy. While forecasting mental health scores alone offers limited insights into the impact of events on community well-being, quasi-experimental designs like the Longitudinal Regression Discontinuity Design (LRDD) from econometrics help researchers derive more effects that are more likely to be causal from observational data. LRDDs aim to extrapolate the size of changes in an outcome (e.g. a discont…

New Pentagon policy undercuts trans troops' ability to ask to stay in the military, AP learns (Konstantin Toropin/Associated Press)
https://apnews.com/article/transgender-troops-ban-military-trump-9c3018034a2608a28f6146fe17f0fe4b
http://www.memeorandum.com/251031/p147#a251031p147

Replaced article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[1/10]:
- Adaptive Policy Learning to Additional Tasks
Wenjian Hao, Zehui Lu, Zihao Liang, Tianyu Zhou, Shaoshuai Mou

The Covid revenge policy (Vox)
https://www.vox.com/today-explained-podcast/459432/covid-vaccine-rfk-trump-mrna
http://www.memeorandum.com/250831/p54#a250831p54

Climate policy should reduce greenhouse gas emissions (Matthew Yglesias/Slow Boring)
https://www.slowboring.com/p/climate-policy-should-reduce-greenhouse
http://www.memeorandum.com/251030/p42#a251030p42

More than 100 judges have ruled against the Trump admin's mandatory detention policy (Kyle Cheney/Politico)
https://www.politico.com/news/2025/10/31/trump-administration-mandatory-detention-deportation-00632086
http://www.memeorandum.com/251031/p125#a251031p125

Trump Is Starting to Turn on the People He Handpicked (Wall Street Journal)
https://www.wsj.com/politics/policy/trump-administration-firings-65d54660
http://www.memeorandum.com/250830/p3#a250830p3

Oz: Coming spike in ACA premium payments 'not the big issue' (Nathaniel Weixel/The Hill)
https://thehill.com/policy/healthcare/5580142-affordable-care-act-premiums-increase-dr-oz/
http://www.memeorandum.com/251030/p45#a251030p45

ICE's forced face scans to verify citizens is unconstitutional, lawmakers say (Ashley Belanger/Ars Technica)
https://arstechnica.com/tech-policy/2025/10/ices-forced-face-scans-to-verify-citizens-is-unconstitutional-lawmakers-say/
http://www.memeorandum.com/251029/p119#a251029p119

Two Virginia school districts sue Education Dept. in fight over gender policies (Karina Elwood/Washington Post)
https://www.washingtonpost.com/education/2025/08/29/virginia-school-district-sues-education-department-transgender-policy/
http://www.memeorandum.com/250829/p43#a250829p43