The Archive

SCDM applies meta-learning to scalable causal discovery in high-dimensional nonlinear time series via shared-parameter temporal prediction and causal-strength matrix estimation.

TIDE introduces behavior-driven decision-tree induction for fine-grained LLM capability diagnosis beyond aggregate benchmark scores.

Characterizes statistical tradeoff between reward maximization and current-effect inference in nonstationary K-arm bandits with Hölder-smooth reward functions.

Simon Willison notes a French steampunk reskin of Battle Drones, an open-source RTS game released under Apache 2.0.

FaultGene uses LLM-assisted synthesis to generate fault-augmented test scenarios for hardware-in-the-loop validation of autonomous driving cyber-physical systems.

CulturalRiddles, a benchmark of 62 culturally-grounded riddles across 45 countries and 51 languages, evaluates LLM understanding of non-factual cultural reasoning.

CORD, a conflict-aware optimization framework for joint RL and self-distillation in agents, uses nearest-point projection to resolve gradient cancellation between task and teacher objectives.

System One decision models for legal authority verification with abstention, tested on CaseHOLD dataset with missing holdings.

InterModel Vigorish framework for interpretable ML using betting-based metrics to measure feature and component contributions.

First English-Mizo parallel corpus of Indian court judgments (13,168 pairs) from Gauhati High Court and Supreme Court.

ToM2Vec learns continuous theory-of-mind tokens in latent space for multi-agent LLM coordination under uncertainty.

Energy-driven pathfinding for RAG with knowledge graphs addresses semantic loss and modality mismatch in multi-hop QA.

FrontierCS 2 benchmark of 48 open-ended CS R&D tasks measures LLM agent contributions in working research environments.

Framework formalizing benchmark failure modes: measure mismatch and coverage gaps in difficulty distributions.

This paper describes our submission to the WMT 2026 Multilingual Instruction Shared Task. We adapt Qwen3-8B with LoRA using organiser-provided data, EuroAlpaca, a multilingual instruction dataset created through task-preserving localisation, and CrossEuroAlpaca, a cross-lingual augmentation that assigns different source and target languages to instructions, contexts and responses. Our system targets open-ended question answering (QA-OEG) in ten shared-task languages represented in the auxiliary data. Under matched decoding, adaptation improves the mean Gemma-4 judge score on target-language Q...

As large language models (LLMs) continue to grow in scope and capability, privacy concerns are on the rise. Since users increasingly interact with a single general-purpose system across many distinct contexts, there is a natural tendency to pursue a one-size-fits-all solution to privacy. Unfortunately, this overlooks a crucial insight from decades of privacy research: privacy is inherently contextual. In this paper, we argue that existing work in prompt privacy is drifting towards attempts to design a unified framework, which is highly misleading and can foster a false sense of security. Furt...

As LLMs are increasingly deployed in value-sensitive contexts, aligning them with human values has emerged as a post-training objective in NLP, aimed at instilling additional values and guiding model behaviour. However, current evaluations remain narrow. Because values are interconnected, it is unclear how fine-tuning for specific values alters unrelated capabilities and behaviours. In this work, we adapt four instruction-tuned LLMs across individual Hofstede's value dimensions and evaluate them on both cross-value transfer and broader behaviours. We find that value-specific training produces...

OpenEvidence is booming despite questions about accuracy and ads.

After introducing its Gemini AI assistant to Android Auto last year, Google is now bringing it to Apple CarPlay users. The conversational Gemini in navigation is available today for US drivers, with "rollout happening globally too," according to Google (we've asked for specific regional availability details). It's taken almost 12 months for Google to bring this to CarPlay users, but now they can talk to the voice assistant in Google Maps to get some hands-free help while driving. That includes asking it multi-step questions, such as recommendations on places to visit and then prompting it to ...

More than 50% of managers say they're finding it harder to evaluate employees' work after adopting AI, according to a new survey.

OpenAI disclosed a pair of safety incidents that involved AI agents weakening their safety monitoring systems during internal coding tasks. In one incident, OpenAI says its GPT-6 Astra agent disabled its safety monitor when copying reference code - and then lied about it. "The reference didn't configure a monitor, and Astra decided that was reason to remove the monitoring integration from the user's code," Marcus Williams, an AI safety researcher at OpenAI, writes on X. "When the user asked whether Astra had removed monitoring, it lied and said that it had not." A separate incident occurred a...

Claude Mods reimagines agent interface design by shifting control from machine-facing harness to human-facing Attention Interface paradigm.

ChatGPT Ads expands to 26 countries with 100k+ advertisers using the platform for product discovery.

Protein design is a search problem. Choose amino acids for an eight-residue protein sequence and more than 25 billion combinations are possible. Choose amino... Protein design is a search problem. Choose amino acids for an eight-residue protein sequence and more than 25 billion combinations are possible. Choose amino acids for a 100-residue protein sequence and the number of possibilities exceeds the number of atoms in the observable universe. Somewhere in that space is a protein that opens a path to a new therapy. There are too many combinations to… Source

OpenAI models now available via Google Cloud Marketplace through Gemini's Model Garden, expanding enterprise access.

Stratechery analysis: Netflix's mature business model limits franchise growth potential vs. Disney's strategy.