safety

DMGD: Train-Free Dataset Distillation with Semantic-Distribution Matching in Diffusion Models

arXiv:2605.03877v1 Announce Type: new Abstract: Dataset distillation enables efficient training by distilling the information of large-scale datasets into significantly smaller synthetic datasets. Dif

Human

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Safety

@Dr_Gingerballs At least actual ponzi schemes don't light their cash on fire... they just cant meet redemptions at the level of their inflat…

DGX agent

@Dr_Gingerballs At least actual ponzi schemes don't light their cash on fire... they just cant meet redemptions at the level of their inflated fake earnings. After all of the hyperscalers burn every l

safetygary-marcus--x

6 May 2026

Efficient Temporal Datalog Materialisation for Composite Event Recognition

arXiv:2605.02488v1 Announce Type: new Abstract: Several applications demand the timely detection of critical situations, such as threats to safety and transparency, over high-velocity streams of symbo

EvoJail: Evolutionary Diverse Jailbreak Prompt Generation for Large Language Models

arXiv:2605.02921v1 Announce Type: cross Abstract: As LLMs continue to shape real-world applications, automated jailbreak generation becomes essential to reveal safety weaknesses and guide model improv

False Friends in the Shell: Unveiling the Emoticon Semantic Confusion in Large Language Models

arXiv:2601.07885v2 Announce Type: replace-cross Abstract: Emoticons are widely used in digital communication to convey affective intent, yet their safety implications for Large Language Models (LLMs)

FIBER: A Differentially Private Optimizer with Filter-Aware Innovation Bias Correction

arXiv:2605.03425v1 Announce Type: new Abstract: Differentially private (DP) training protects individual examples by adding noise to gradients, but the injected noise interacts nontrivially with adapt

FINER-SQL: Boosting Small Language Models for Text-to-SQL

arXiv:2605.03465v1 Announce Type: cross Abstract: Large language models have driven major advances in Text-to-SQL generation. However, they suffer from high computational cost, long latency, and data

FORMULA: FORmation MPC with neUral barrier Learning for safety Assurance

arXiv:2604.04409v2 Announce Type: replace Abstract: Multi-robot systems (MRS) are essential for large-scale applications such as disaster response, material transport, and warehouse logistics, yet ens

From SFT to RL: Demystifying the Post-Training Pipeline for LLM-based Vulnerability Detection

arXiv:2602.14012v2 Announce Type: replace-cross Abstract: The integration of LLMs into vulnerability detection (VD) has shifted the field toward more interpretable and context-aware analysis. While po

@GaryMarcus I can't believe this is still a thing that 'experts' haven't caught up with. @GaryMarcus has been saying this forever, and those…

@GaryMarcus I can't believe this is still a thing that 'experts' haven't caught up with. @GaryMarcus has been saying this forever, and those of us who have actually dug into the tech, analyzed it, use

Geometry Forcing: Marrying Video Diffusion and 3D Representation for Consistent World Modeling

arXiv:2507.07982v2 Announce Type: replace Abstract: Videos inherently represent 2D projections of a dynamic 3D world. However, our analysis suggests that video diffusion models trained solely on raw v

Global and Local Topology-Aware Attention with Persistent Homology and Euler Biases for Time-Series Forecasting

arXiv:2605.03163v1 Announce Type: new Abstract: Scientific time series often encode predictive geometric structure, including connectivity, cycles, shell-like geometry, directional changes, and nonlin

Google, Microsoft and xAI agree to allow government safety checks of their AI models prior to release

Google LLC, Microsoft Corp. and xAI have agreed to share unreleased versions of their artificial intelligence models with the U.S. Department of Commerce to ensure the technologies do not pose a threa

safetysiliconangle

Governing What the EU AI Act Excludes: Accountability for Autonomous AI Agents in Smart City Critical Infrastructure

arXiv:2605.01091v1 Announce Type: cross Abstract: When a traffic signal controller adjusts green phases and a grid manager curtails power on the same corridor, each system may comply with its own obli

GRAFT: Auditing Graph Neural Networks via Global Feature Attribution

arXiv:2605.03377v1 Announce Type: new Abstract: Graph Neural Networks (GNNs) achieve strong performance on node classification tasks but remain difficult to interpret, particularly with respect to whi

Grounding Multi-Hop Reasoning in Structural Causal Models via Group Relative Policy Optimization

arXiv:2605.01482v1 Announce Type: new Abstract: Multi-Hop Fact Verification (MHFV) necessitates complex reasoning across disparate evidence, posing significant challenges for Large Language Models (LL

GRPO-TTA: Test-Time Visual Tuning for Vision-Language Models via GRPO-Driven Reinforcement Learning

arXiv:2605.03403v1 Announce Type: new Abstract: Group Relative Policy Optimization (GRPO) has recently shown strong performance in post-training large language models and vision-language models. It ra

Healthcare AI GYM for Medical Agents

arXiv:2605.02943v1 Announce Type: new Abstract: Clinical reasoning demands multi-step interactions -- gathering patient history, ordering tests, interpreting results, and making safe treatment decisio

Height Control and Optimal Torque Planning for Jumping With Wheeled-Bipedal Robots

arXiv:2605.03302v1 Announce Type: new Abstract: This paper mainly studies the accurate height jumping control of wheeled-bipedal robots based on torque planning and energy consumption optimization. Du

Heterogeneous Graph Importance Scoring and Clustering with Automated LLM-based Interpretation

arXiv:2605.02919v1 Announce Type: new Abstract: Urban bridge networks are critical infrastructure whose disruption can cascade into severe impacts on transportation, emergency services, and economic a

HiMAC: Hierarchical Macro-Micro Learning for Long-Horizon LLM Agents

arXiv:2603.00977v2 Announce Type: replace-cross Abstract: Large language model (LLM) agents have recently demonstrated strong capabilities in interactive decision-making, yet they remain fundamentally

How Sam ('You parachute him onto a cannibal island, and he comes back five years later as king”) Altman operates:

How Sam ('You parachute him onto a cannibal island, and he comes back five years later as king”) Altman operates: counsel: 'by fall of 2023 did you percieve altman was not candid with you? truthful? h

Human-in-the-Loop Uncertainty Analysis in Self-Adaptive Robots Using LLMs

arXiv:2605.02983v1 Announce Type: new Abstract: Self-adaptive robots operate in dynamic, unpredictable environments where unaddressed uncertainties can lead to safety violations and operational failur

I repeat, the bubble is in the 'e' not the 'p' in today's PE ratios. The hucksters and talking heads will, as always, fail to realize until …

I repeat, the bubble is in the 'e' not the 'p' in today's PE ratios. The hucksters and talking heads will, as always, fail to realize until it's too late. But it's a very simple set up. Hyperscalers g

If Forbes had only waited to hear the testimony at this week’s trial Or read @_KarenHao’s book Or @RonanFarrow’s @newyorker investigation Or…

If Forbes had only waited to hear the testimony at this week’s trial Or read @_KarenHao’s book Or @RonanFarrow’s @newyorker investigation Or my own writings since fall 2023 They would have realized ho

Important nuance: the jury at this Musk-OpenAI trial is an advisory jury, hence not binding on the judge, and is only looking at liability…

Important nuance: the jury at this Musk-OpenAI trial is an *advisory* jury, hence not binding on the judge, and is only looking at liability (not damages, if any). Thanks to @bahhradx for correcting a

Instance-Level Costs for Nuanced Classifier Evaluation

arXiv:2605.03135v1 Announce Type: new Abstract: Standard classification treats all errors equally, but in content moderation, medical screening, and safety-critical applications, mistakes on clear-cut

Intervention Complexity as a Canonical Reward and a Measure of Intelligence

arXiv:2605.02175v1 Announce Type: new Abstract: The Legg--Hutter universal intelligence measure provides a rigorous scalar assessment of general intelligence as expected reward across all computable e

just want to go back to how much intense pushback i got on this story at all levels of the company at the time, and how people speak very di…

just want to go back to how much intense pushback i got on this story at all levels of the company at the time, and how people speak very differently when under the threat of perjury lot of names etch

Khala: Scaling Acoustic Token Language Models Toward High-Fidelity Music Generation

arXiv:2605.01790v1 Announce Type: cross Abstract: A common design pattern in high-quality music generation is to handle structure and fidelity in different representation spaces: a generator first mod

Large Language Models are Universal Reasoners for Visual Generation

arXiv:2605.04040v1 Announce Type: new Abstract: Text-to-image generation has advanced rapidly with diffusion models, progressing from CLIP and T5 conditioning to unified systems where a single LLM bac

Learning Reactive Dexterous Grasping via Hierarchical Task-Space RL Planning and Joint-Space QP Control

arXiv:2605.03363v1 Announce Type: new Abstract: In this work, we propose a hybrid hierarchical control framework for reactive dexterous grasping that explicitly decouples high-level spatial intent fro

Like tricksters, LLMs have perfected the art of plausibility, says Tim Harford: https://ft.trib.al/5Foo2YD

Tim Harford compares large language models to tricksters, arguing that LLMs excel at generating plausible-sounding text without necessarily ensuring accuracy or truthfulness. The article likely explor

LLM-XTM: Enhancing Cross-Lingual Topic Models with Large Language Models

arXiv:2605.03299v1 Announce Type: new Abstract: Cross-lingual topic modeling aims to discover shared semantic structures across languages, yet existing models depend on sparse bilingual resources and

Logic-Constrained Shortest Paths for Flight Planning

arXiv:2412.13235v4 Announce Type: replace Abstract: The logic-constrained shortest path problem (LCSPP) combines a one-to-one shortest path problem with satisfiability constraints imposed on the routi

MAGE: Safeguarding LLM Agents against Long-Horizon Threats via Shadow Memory

arXiv:2605.03228v1 Announce Type: cross Abstract: As large language model (LLM)-powered agents are increasingly deployed to perform complex, real-world tasks, they face a growing class of attacks that

Many trials feel like Rashomon with different witnesses. The amazing thing about Musk-OpenAI is how much agreement there has been (at least …

Many trials feel like Rashomon with different witnesses. The amazing thing about Musk-OpenAI is how much agreement there has been (at least so far) on the facts. The question is really whether what Op

Memorization In Stable Diffusion Is Unexpectedly Driven by CLIP Embeddings

arXiv:2605.02908v1 Announce Type: new Abstract: Understanding how textual embeddings contribute to memorization in text-to-image diffusion models is crucial for both interpretability and safety. This

MILD: Mediator Agent System with Bidirectional Perception and Multi-Layered Alignment for Human-Vehicle Collaboration

arXiv:2605.01507v1 Announce Type: new Abstract: Prior studies report that partial driving automation can increase the cognitive demands on human drivers. This effect largely arises from human drivers'

MINT: Minimal Information Neuro-Symbolic Tree for Objective-Driven Knowledge-Gap Reasoning and Active Elicitation

arXiv:2602.05048v2 Announce Type: replace Abstract: Joint planning through language-based interactions is a key area of human-AI teaming. Planning problems in the open world often involve various aspe

Mira Murati tells the court that she couldn’t trust Sam Altman’s words

Mira Murati, OpenAI's former CTO, has testified under oath that CEO Sam Altman lied to her about the safety standards for a new AI model. In a video deposition shown during the ongoing Musk v. Altman

safetythe-verge-ai

Mira Murati’s testimony is gripping – and what it makes absolutely clear is how utterly wrong most of Twitter was about why Sam was fired. –…

Mira Murati’s testimony is gripping – and what it makes absolutely clear is how utterly wrong most of Twitter was about why Sam was fired. – It had nothing per se to do with AI safety - It had nothing

Mix3R: Mixing Feed-forward Reconstruction and Generative 3D Priors for Joint Multi-view Aligned 3D Reconstruction and Pose Estimation

arXiv:2605.03359v1 Announce Type: new Abstract: Recent trends in sparse-view 3D reconstruction have taken two different paths: feed-forward reconstruction that predicts pixel-aligned point maps withou

Model Routing as a Trust Problem: Route Receipts for Adaptive AI Systems

arXiv:2605.01710v1 Announce Type: new Abstract: AI products often route requests through version aliases, service tiers, tool choices, regional endpoints, fallback rules, or safety handling before res

Model Spec Midtraining: Improving How Alignment Training Generalizes

arXiv:2605.02087v1 Announce Type: new Abstract: Some frontier AI developers aim to align language models to a Model Spec or Constitution that describes the intended model behavior. However, standard a

Multilingual Safety Alignment via Self-Distillation

arXiv:2605.02971v1 Announce Type: cross Abstract: Large language models (LLMs) exhibit severe multilingual safety misalignment: they possess strong safeguards in high-resource languages but remain hig

Musk v. Altman: Mira Murati testifies that Sam Altman lied to her about the safety standards for a new OpenAI model and that he made her work more difficult (Jay Peters/The Verge)

Jay Peters / The Verge: Musk v. Altman: Mira Murati testifies that Sam Altman lied to her about the safety standards for a new OpenAI model and that he made her work more difficult — OpenAI's former C

safetytechmeme

Neuron-Anchored Rule Extraction for Large Language Models via Contrastive Hierarchical Ablation

arXiv:2605.03058v1 Announce Type: new Abstract: A key goal of explainable AI (XAI) is to express the decision logic of large language models (LLMs) in symbolic form and link it to internal mechanisms.