Tootfinder

Some good venting by Steve Klabnik about the sorry state of significant chunks of the AI debate today:
"What is breaking my brain a little bit is that all of the discussion online around AI is so incredibly polarized. This isn’t a “the middle is always right” sort of thing either, to be clear. It’s more that both the pro-AI and anti-AI sides are loudly proclaiming things that are pretty trivially verifiable as not true."

Bridging Offline and Online Reinforcement Learning for LLMs
Jack Lanchantin, Angelica Chen, Janice Lan, Xian Li, Swarnadeep Saha, Tianlu Wang, Jing Xu, Ping Yu, Weizhe Yuan, Jason E Weston, Sainbayar Sukhbaatar, Ilia Kulikov
https://arxiv.org/abs/2506.21495 https://arxiv.org/pdf/2506.21495 https://arxiv.org/html/2506.21495
arXiv:2506.21495v1 Announce Type: new
Abstract: We investigate the effectiveness of reinforcement learning methods for finetuning large language models when transitioning from offline to semi-online to fully online regimes for both verifiable and non-verifiable tasks. Our experiments cover training on verifiable math as well as non-verifiable instruction following with a set of benchmark evaluations for both. Across these settings, we extensively compare online and semi-online Direct Preference Optimization and Group Reward Policy Optimization objectives, and surprisingly find similar performance and convergence between these variants, which all strongly outperform offline methods. We provide a detailed analysis of the training dynamics and hyperparameter selection strategies to achieve optimal results. Finally, we show that multi-tasking with verifiable and non-verifiable rewards jointly yields improved performance across both task types.
toXiv_bot_toot

Cryptographic Data Exchange for Nuclear Warheads
Neil Perry, Daniil Zhukov
https://arxiv.org/abs/2507.20074 https://arxiv.org/pdf/2507.20074

Cryptographic Data Exchange for Nuclear Warheads
Nuclear arms control treaties have historically focused on strategic nuclear delivery systems, leaving nuclear warheads outside formal verification frameworks. This paper presents a cryptographic protocol for secure and verifiable warhead tracking, addressing challenges in nuclear warhead verification without requiring intrusive physical inspections. Our system leverages commitment schemes and zero-knowledge succinct non-interactive arguments of knowledge (zkSNARKs) to ensure compliance with tr…

Improving Neural Network Training using Dynamic Learning Rate Schedule for PINNs and Image Classification
D. Veerababu, Ashwin A. Raikar, Prasanta K. Ghosh
https://arxiv.org/abs/2507.21749

Improving Neural Network Training using Dynamic Learning Rate Schedule for PINNs and Image Classification
Training neural networks can be challenging, especially as the complexity of the problem increases. Despite using wider or deeper networks, training them can be a tedious process, especially if a wrong choice of the hyperparameter is made. The learning rate is one of such crucial hyperparameters, which is usually kept static during the training process. Learning dynamics in complex systems often requires a more adaptive approach to the learning rate. This adaptability becomes crucial to effecti…

Deutschland Hochburg bei E-Bikes in Europa – die Preise sinken
Nirgendwo in Europa wird so viel Geld mit E-Bikes gemacht. Zwar gab es zuletzt einen Dämpfer im Geschäft. Doch langfristig dürfte kein Weg daran vorbeiführen.

Unlocking my own understanding of and ability to build #Swift macros feels like a superpower.
…something something great responsibility, though.
Synthesizing boilerplate and statically-verifiable elements like custom function calls based on macro input… is magic—the good kind.
`@GET("/logs/{userId}/{timing}")`
↘️

Latent Representations for Control Design with Provable Stability and Safety Guarantees
Paul Lutkus, Kaiyuan Wang, Lars Lindemann, Stephen Tu
https://arxiv.org/abs/2505.23210

Latent Representations for Control Design with Provable Stability and Safety Guarantees
We initiate a formal study on the use of low-dimensional latent representations of dynamical systems for verifiable control synthesis. Our main goal is to enable the application of verification techniques -- such as Lyapunov or barrier functions -- that might otherwise be computationally prohibitive when applied directly to the full state representation. Towards this goal, we first provide dynamics-aware approximate conjugacy conditions which formalize the notion of reconstruction error necessa…

Towards Multi-Agent Economies: Enhancing the A2A Protocol with Ledger-Anchored Identities and x402 Micropayments for AI Agents
Awid Vaziry, Sandro Rodriguez Garzon, Axel K\"upper
https://arxiv.org/abs/2507.19550

Towards Multi-Agent Economies: Enhancing the A2A Protocol with Ledger-Anchored Identities and x402 Micropayments for AI Agents
This research article presents a novel architecture to empower multi-agent economies by addressing two critical limitations of the emerging Agent2Agent (A2A) communication protocol: decentralized agent discoverability and agent-to-agent micropayments. By integrating distributed ledger technology (DLT), this architecture enables tamper-proof, on-chain publishing of AgentCards as smart contracts, providing secure and verifiable agent identities. The architecture further extends A2A with the x402 …

VeriThoughts: Enabling Automated Verilog Code Generation using Reasoning and Formal Verification
Patrick Yubeaton, Andre Nakkab, Weihua Xiao, Luca Collini, Ramesh Karri, Chinmay Hegde, Siddharth Garg
https://arxiv.org/abs/2505.20302

VeriThoughts: Enabling Automated Verilog Code Generation using Reasoning and Formal Verification
This paper introduces VeriThoughts, a novel dataset designed for reasoning-based Verilog code generation. We establish a new benchmark framework grounded in formal verification methods to evaluate the quality and correctness of generated hardware descriptions. Additionally, we present a suite of specialized small-scale models optimized specifically for Verilog generation. Our work addresses the growing need for automated hardware design tools that can produce verifiably correct implementations …

When Life Gives You Samples: The Benefits of Scaling up Inference Compute for Multilingual LLMs
Ammar Khairi, Daniel D'souza, Ye Shen, Julia Kreutzer, Sara Hooker
https://arxiv.org/abs/2506.20544

When Life Gives You Samples: The Benefits of Scaling up Inference Compute for Multilingual LLMs
Recent advancements in large language models (LLMs) have shifted focus toward scaling inference-time compute, improving performance without retraining the model. A common approach is to sample multiple outputs in parallel, and select one of these as the final output. However, work to date has focused on English and a handful of domains such as math and code. In contrast, we are most interested in techniques that generalize across open-ended tasks, formally verifiable tasks, and across languages…