Safe Exploration via Policy Priors
arXiv:2601.19612v4 Announce Type: replace-cross Abstract: Safe exploration is a key requirement for reinforcement learning (RL) agents to learn and adapt online, beyond controlled (e.g. simulated) environments.…
Full feed
Feed 27,607 posts
Every story across every category, newest first. Each card links to the original publisher; daily-brief posts open as editorial pages.
Position: The Alignment Community is Unintentionally Building a Censor's Toolkit
arXiv:2608.12346v1 Announce Type: new Abstract: This position paper argues that modern AI alignment methods - originally designed to prevent harmful output - are dual-use technologies that…
Agreement Is Not Alignment: Divergent Moral Grounds in Human and LLM Ethical Judgments
arXiv:2608.12368v1 Announce Type: new Abstract: Agreement with human judgments is a common proxy for evaluating the alignment of large language models (LLMs). Yet agreement in final…
Multi-Agent Scheduling with LLM-Assisted Contract Net Negotiation for Stream Processing in Mobile Edge Computing
arXiv:2608.12371v1 Announce Type: new Abstract: Stream-processing systems increasingly operate across heterogeneous mobile edge--cloud infrastructures, where workload volatility, resource contention, and stringent quality-of-service (QoS) requirements complicate decentralized…
Don't Want Your LLM to Recommend Nuclear Strike? Try Asking It in Japanese
arXiv:2608.12373v1 Announce Type: new Abstract: Large language models are increasingly used in strategic and advisory contexts, yet their safety alignment is typically evaluated in English only.…
Position: We Need Practical AI Alignment Methods to Mirror Human Reasoning
arXiv:2608.12372v1 Announce Type: new Abstract: AI systems are increasingly employed as decision aids, decision delegates, or autonomous decision-makers. This position paper argues that in many settings,…
Coordinated incentives in AI-generated misinformation governance
arXiv:2608.07070v2 Announce Type: replace-cross Abstract: With the rapid diffusion of AI-generated content, AI-driven misinformation is becoming increasingly pervasive and difficult to govern, undermining information credibility and…
Dual-Flow Transformers: Decoupling the Primary Prefill Path from Additional Decode Computation
arXiv:2608.12385v1 Announce Type: new Abstract: As large language models serve more requests, cumulative inference cost is becoming increasingly important relative to one-time training cost. The two…
Into the ORBIT for Time Series: Training Regimes for Foundation Models
arXiv:2608.13262v1 Announce Type: cross Abstract: Time series foundation models (TSFMs) have advanced primarily through architectural innovation, while training regimes for large-scale heterogeneous corpora remain under-explored. As…
Operationalizing Cyber Threat Intelligence with GraphRAG
arXiv:2608.13050v1 Announce Type: cross Abstract: When a security researcher publishes a report on a cyberattack, detection engineers are supposed to turn it into working detection rules.…
Identification of Probabilities of Causation: from Recursive to Closed-Form Bounds
arXiv:2505.15274v4 Announce Type: replace Abstract: Probabilities of causation (PoCs) are fundamental quantities for counterfactual analysis and personalized decision making. However, existing analytical results are largely confined…
UniTexture: Cross-Task Universal Adversarial Textures for Vision-Language-Action Models
arXiv:2608.13453v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models have emerged as generalist robotic policies capable of following diverse language instructions and performing a wide range of…
Yes, Q-learning Helps Offline In-Context RL
arXiv:2502.17666v5 Announce Type: replace-cross Abstract: Existing offline in-context reinforcement learning (ICRL) methods have predominantly relied on supervised training objectives, which are known to have limitations in…
This just means nobody else was even running an honest business and their open sourcing has made no difference. Why the hell is the model developer th…
This just means nobody else was even running an honest business and their open sourcing has made no difference. Why the hell is the model developer…
wild to see Sol at 14x speed:
wild to see Sol at 14x speed:OpenAI: Previewing Ultrafast mode: GPT-5.6 Sol at up to 14x the speed.Launching first in the OpenAI API to a select…
chatgpt for saving you money
chatgpt for saving you moneyjason: has chatgpt or codex ever saved you actual money?caught an incorrect invoice, found a billing mistake, canceled subscriptions, negotiated a bill,…
1 week to GLM-5.3 it seems
1 week to GLM-5.3 it seemsjietang: sooooooon
it's unusable the model is obsessed with having NATIVE VISION, it is going psychotic without eyes. Would have probably aced a very complex multi-hour …
it's unusablethe model is obsessed with having NATIVE VISION, it is going psychotic without eyes. Would have probably aced a very complex multi-hour workload if not…
GitHub Copilot weekly releases — August 10
From new models and portable plugins to smoother agent workflows, this week’s updates make GitHub Copilot more flexible across editors, the command line, and the Copilot…
/ultrafast
/ultrafastOpenAI: Previewing Ultrafast mode: GPT-5.6 Sol at up to 14x the speed.Launching first in the OpenAI API to a select group of customers with expanded access…
Create a Reasoning-Focused LLM: A Practical Guide to Streaming, Curating, and Fine-Tuning the SupraLabs Reasoning Corpus
This tutorial provides a complete workflow for building a compact, reasoning-focused language model. By streaming the SupraLabs reasoning corpus from Hugging Face, we apply quality filters…
Lmao. what a retarded system, not even a pagerank type correction don't piss any group of people off I guess unless you have a rabid audience, a small…
Lmao. what a retarded system, not even a pagerank type correctiondon't piss any group of people off I guessunless you have a rabid audience, a small…
DarwinX: Evolving Agent Harnesses Through Natural Selection
An LLM agent's capability depends not only on model weights but on its harness: prompts, tools, skills, and control flow. Self-improvement loops already edit harnesses, yet…
Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus
We present the first systematic study of Massive activations (MAs) in layer-interleaved HLA LLMs and uncover two architecture-aligned morphologies: MAs consistently spike immediately before full attention…