A-Evolve-Training
“We report an autonomous system that runs this loop with no human in the loop, post-training a 30B Nemotron across four rounds over multiple weeks.”
The authors, in the abstract
Autonomous Post-Training of a 30B Model
AI Research Radar
June 20–26: 10 frontier papers out of 1,973 read.
Frontier
Papers rated as moving the field
How we chooseHOLMES
“Experiments show that current LLMs still struggle on HOLMES, with an average accuracy of only 50.64% and the best model reaching 59.54%.”
OTTER
“Evaluated on 457 AdvBench prompts across four GPT models, OTTER raises average ASR from 7.0% to 84.0%.”
Safety in Self-Evolving LLM Agent Systems
“Comparative case studies of two open-source frameworks demonstrate that evolution-native design activates 3.5× more attack surface cells and achieves a 100% attack persistence rate (40/40 payloads across all CIA+Privacy categories), while co-located security scanners block only 2.5% of attacks.”
Know2Guess
From Sounds to Scenes
“Experiments across multiple LALMs demonstrate that effective auditory scene understanding requires integration over all auditory layers, rather than reliance on speech or sound alone, underscoring the necessity of CASU for advancing complex audio understanding in LALMs.”
Qwen-AgentWorld
“Empirical results demonstrate that Qwen-AgentWorld significantly outperforms existing frontier models.”
Inherited Circuits, Learned Semantics
Vesta
BioMatrix
The week by field
Jun 20–26: 164 cleared, 10 frontier
The week’s papers have not been sorted into fields yet.
Models under test
Papers this week that name the model
Labs publishing
Papers this week with an author at the lab
Also cleared
154 papers
- HarmVideoBench: Benchmarking Harmful Video Understanding in Large Multimodal ModelsByteDance, Tencent
- JetSpec: Breaking the Scaling Ceiling of Speculative Decoding with Parallel Tree DraftingStepFunCode
- Improved Large Language Diffusion ModelsByteDanceCode
- PolicyAlign: Direct Policy-Based Safety Alignment for Large Language ModelsAlibabaCode
- Specialize Roles, Mix Deployments: Pushing the Cost-Accuracy Frontier of LLM Agent TeamsMicrosoftCode
- Tmax: A simple recipe for terminal agentsAllenAICode
- Discovering Millions of Interpretable Features with Sparse AutoencodersAlibaba
- Scalable Behavior Cloning with Open Data, Training, and EvaluationAmazon
- PhoneBuddy: Training Open Models for Agentic Phone UseTencent
- Power-Flexible AI Data Centers: A New Paradigm for Grid-Responsive ComputeNVIDIA
- Quantization Inflates Reasoning: Token Inflation as a Hidden Cost of Low-Bit Reasoning ModelsMicrosoft
- Tracing Target Answers in Poisoned Retrieval Corpora via Token Influence AttributionIBM
All 154 papers
- Autodata: An agentic data scientist to create high quality synthetic dataMeta
- Self-Recognition Finetuning can Prevent and Reverse Emergent MisalignmentGoogle
- AgenticOS: An Intent-Oriented Secure Operating System Architecture for Autonomous AI AgentsTencent
- Warning labels shift perceptions of sycophantic AI, but not its influenceMicrosoft
- Investigating The Security of Modern AI and Cloud InfrastructureNVIDIA
- Understanding the (In)Security of Vibe-Coded ApplicationsMicrosoft
- Evaluation Awareness Is Not One Capability: Evidence from Open Language ModelsMicrosoft
- AI Exposure Scores: what they measure, what they miss, and what comes nextCohere
- LIBERO-Safety: A Comprehensive Benchmark for Physical and Semantic Safety in Vision-Language-Action ModelsMicrosoft
- Do Safety Guardrails Need to Reason? LeanGuard: A Fast and Light Approach for Robust ModerationCode
- Prompt Injection in Automated Résumé Screening with Large Language Models: Single and Multi-Injection SettingsCode
- SingGuard: A Policy-Adaptive Multimodal LLM Guardrail with Dynamic ReasoningCode
- Hitting a Moving Target: Test-Time Adaptation for AI Text Detection under Continual Distribution ShiftCode
- REALM: A Unified Red-Teaming Benchmark for Physical-World VLMsCode
- AutoDojo: Adaptive Black-Box Attacks Reveal the Limits of IPI Defenses and Task-Specification Effects in LLM AgentsCode
- Think Twice Before You Act: Protecting LLM Agents Against Tool Description Poisoning via Isolated PlanningCode
- Harness-MU: A Safe, Governed, and Effective Harness for Multi-User LLM AgentsCode
- AgentLens: Interpretable Safety Steering via Mechanistic Subspaces for Multi-Turn Coding AgentCode
- Benchmarking Open-Weight Foundation Models for Global AI Technical Governance
- Dream machine -- the next creative economy
- Divergent Recommendations, Convergent Diagnoses: Cross-Provider Failure-Mode Convergence in AI Commercial Recommendation
- The Governance Inversion Hypothesis: Why More AI Regulation May Produce Less Organisational Control
- Necessary but Not Sufficient: Temperature Control and Reproducibility in LLM-as-Judge Safety Evaluations
- TEMPO-Diffusion: Temporally Exposed Malicious Poisoning of Diffusion Models
- Governing Actions, Not Agents: Institutional Attestation as a Governance Model for Autonomous AI Systems
- Instruction Bleed: Cross-Module Interference in Prompt-Composed Agentic Systems
- When Agents Meet Electric Bus Fleet Operations: Pricing Behavior, Trade-offs, and Policy Implications in an Aggregator Framework
- Otter Weather: Skillful and Computationally Efficient Medium-Range Weather Forecasting
- VIGIL: Runtime Enforcement of Behavioral Specifications in AI Agent Skills
- Tactile-WAM: Touch-Aware World Action Model with Tactile Asymmetric Attention
- MIRROR: Novelty-Constrained Memory-Guided MCTS Red-Teaming for Agentic RAG
- The Capability Frontier: Benchmarks Miss 82% of Model Performance
- Confidence-Aware Tool Orchestration for Robust Video Understanding
- Chai: Agentic Discovery of Cryptographic Misuse Vulnerabilities
- Auditing Framing-Sensitive Behavioral Instability in Large Language Models for Mental Health Interactions
- ShareLock: A Stealthy Multi-Tool Threshold Poisoning Attack Against MCP
- Learning to Fold: prizewinning solution at LeHome Challenge 2026 (1st place online, 2nd offline)
- From Celebrities to Anyone: Characterizing AI Nudification Content, Technology, and Community Dynamics on 4chan
- Small edits, large models: How Wikipedia advocacy shapes LLM values
- Do Thinking Tokens Help with Safety?
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- Forget to Improve: On-Device LLM-Agent Continual Learning via Budget-Curated Memory
- Transferable Attack against Face Swapping in an Extended Space
- Generative AI for Safe and Photorealistic Drone Light Shows
- How Reliable Is Your Jailbreak Judge? Calibration and Adversarial Robustness of Automated ASR Scoring
- Concept Removal for Frontier Image Generative Models
- Leaking Circuit Secrets: Gradient Leakage Attacks on Graph Neural Networks
- SurgAtlas: A Large-Scale Surgical Video-Language Dataset with 2,391 Hours of Open and Minimally Invasive Surgery
- Privacy Vulnerabilities of Attention Layers in Tabular Foundation Models and Protection of High-Risk Queries
- Detect, Unlearn, Restore: Defending Text Summarization Models Against Data Poisoning
- The Unfireable Safety Kernel: Execution-Time AI Alignment for AI Agents and Other Escapable AI Systems
- Real-Time Voice AI Hears but Does Not Listen
- Do Foundation Models See Biology? Evaluating Attention Coherence with Spatial Transcriptomics in Glioblastoma
- How Much Can We Trust LLM Search Agents? Measuring Endorsement Vulnerability to Web Content Manipulation
- One Year Later...The Harms Persist, But So Do We!
- RIFT-Bench: Dynamic Red-teaming For Agentic AI Systems
- Cyclic Denoising Reveals Ultrastable Memories in Diffusion Models
- AutoSpec: Safety Rule Evolution for LLM Agents via Inductive Logic Programming
- Securing LLM-Agent Long-Term Memory Against Poisoning: Non-Malleable, Origin-Bound Authority with Machine-Checked Guarantees
- When Helpfulness Overrides Causal Caution: Context-Dependent Suppression and Recovery in LLMs
- Age of LLM: A Strategic 1v1 Benchmark for Reasoning, Diplomacy and Reliability of Large Language Models under Fog of War
- Poisoned Playbooks: Demystifying Knowledge Poisoning Effects on AI Security Agents
- Detecting AI Coding Agents in Open Source: A Validated Multi-Method Census of 180 Million Repositories
- S1-Omni-Image: A Unified Model for Scientific Image Understanding, Generation, and Editing
- The African Language Tax: Quantifying the Cost, Latency, and Context Penalty of Tokenizing African Languages in Frontier LLMs
- Red-Teaming the Agentic Red-Team
- OpenThoughts-Agent: Data Recipes for Agentic Models
- BenchX: Benchmarking AI Models for Cancer Detection and Localization with Demographic and Protocol Biases
- InSight: Self-Guided Skill Acquisition via Steerable VLAs
- From Untrusted Input to Trusted Memory: A Systematic Study of Memory Poisoning Attacks in LLM Agents
- MalSkillBench: A Runtime-Verified Benchmark of Malicious Agent Skills
- TinyGiantALM: A Compact Audio-Language Model for Intent-Aware Reasoning under Resource Constraints
- What the Eyes See, the LLMs Miss: Exploiting Human Perception for Adversarial Text Attacks
- Mind your key: An Empirical Study of LLM API Credential Leakage in iOS Apps
- One Goal, Many Commands: Characterizing Denylist Fragility in AI Agents
- Do Activation Monitors Survive Model Updates? Benchmarking, Predicting, and Repairing Activation-Monitor Staleness
- Ghost Vectors: Soft-Deleted Embeddings Remain Reconstructible in HNSW Vector Databases
- Infrastructure for the Agentic Web: Gap Analysis and Architecture from the Agentverse Platform
- Human-Less LLM Serving: Quantifying the Human Tax on Throughput
- Signals in the Noise: Open Source Intelligence (OSINT) for AI Loss of Control Detection
- BELLS-O: Evaluating the Operational Trade-offs of LLM Supervision Systems
- Agent Behavior Mining: Generative AI Agent Governance in Business Processes
- SafeDojo: Safe Reinforcement Learning for VLA via Interactive World Model
- FairTutor: Equity-Aware Pedagogical LLM Routing for Budget-Constrained AI Tutoring
- MIRAGE: Stealthy Visual Prompt Injection for Vulnerability Detection in Web Agents
- Mirage: a Clean-Label Backdoor against LiDAR 3D Object Detection
- Co-Construction Blindness and Asymmetric Epistemic Vulnerability in Human-LLM Interaction
- FirstPass: Grounding AI Scientific Judgment in Multi-Round Editorial Outcomes
- The Substrate Collapse: AI Code Generation Invalidates Authorship-Based Knowledge Metrics
- Whose Agent Are You? Multi-Layer Fingerprinting and Attribution of Autonomous Web Agents
- Generative Responsible AI Data Evaluation Schema (GRAIDES) for AI Assurance in Local Government
- Demystifying Numerical Instability in LLM Inference: Achieving Reproducible Inference for Mission-Critical Tasks with HEAL
- Local LLM Agents as Vulnerable Runtimes:A Source-Code Audit of the Agent Runtime Layer
- MammoExpert: Benchmarking Chain-of-Thought Reasoning in Mammography Diagnosis
- Answer Engineering: Local Trajectory Editing for Protocol-Constrained Decision Making in Large Language Models
- PulseCX: Breaking the Closed-World Assumption in Real-Time CX
- ChronoLock: Protecting Videos from Unauthorized Text-to-Video Personalization
- AOR-Bench: Do Large Audio Language Models Over-Refuse Pseudo-Harmful Queries?
- BadDreamer: Transferable Backdoor Attacks against Video World Models for Autonomous Driving
- MEDLAYXPLAIN: Benchmarking the Expert-Lay Gap in Medical Vision-Language Models
- Recency/Frequency Adaptive KV Caching for Large Language Model Serving
- Human-AI Interaction Requirements in Public Sector Procurements
- "What Happens Locally, Leaks Globally": Detecting Privacy Leakage Risks in MCP Servers
- Don't Blindly Trust It: How Unreliable Feedback Breaks Tool-Using LLM Agents
- Per-Entity Bias Mapping for AI Visibility: Why Brand Mentions Require Entity-Specific Calibration
- Toward Open Weight Models Without Risks: Separating Public and Private Capabilities in LLMs
- Safe to Check, Unsafe to Use: Relinking at the Compression Boundary of LLM Agents
- Bridging the Age Gap: Towards Detecting Neural Audio Codec Synthesized Elderly Speech Deepfake
- Fixed RAG Compression Collapses Measured Reader Scaling
- Agent-Assisted Side-Channel Attacks on Non-Prefix KV Cache in RAG
- The Language-Energy Divide: Measuring Energy Costs of Multilingual LLM Inference
- AgentRiskBOM: A Risk-Scoping Security Bill of Materials for Agentic AI Systems
- Old Fictions, New Skins: Evaluating the Manipulative Capabilities of LLMs in Healthcare
- L20-Edu-135M: An Auditable Single-GPU Study of Data-Efficient Small Language Modeling
- Revelio: Cost-Efficient Agentic Memory Safety Vulnerability Detection For Repository-Scale Codebases
- T-IMPACT: A Severity-Aware Benchmark for Contextual Image-Text Manipulation
- Formal-Method-Guided Vibe Coding: Closing the Verification Loop on AI-Generated Safety-Critical Software Through Model-Driven Engineering
- Music Playlist Captioning at Scale with Large Language Models
- RAVEN: Agentic RAG for Automated Vulnerability Repair
- Skin-Deep: A Geometric Diagnostic for Alignment Fragility in Large Language Model Representations
- Black-Box Forensics for Conversational LLM Agents
- Catching Lies Without Sending the Video: Privacy-Preserving Multimodal Deception Detection
- One-Prompt Censorship Evasion via Generative Diffusion Models
- Habituation at the Gate: Rising Approval and Declining Scrutiny in Human Review of AI Agent Code
- Language-Specific Sentiment Polarity Biases in Encoder and Large Language Model Classification of Product Reviews
- Towards Robust Personalized Federated Learning: Vulnerability Assessment and Defense Co-Design
- Cloak: Zero-Shot Cross-Embodiment Manipulation by Masking the End-Effector from the VLA
- RLM-Cascade: Response-Level Speculative Decoding for Cost-Efficient LLM API Serving
- The Unseen Hand: Manipulating Model Fairness and SHAP with Targeted Identity Re-Association Attacks
- FedOT: Ownership Verification and Leakage Tracing via Watermarks for Federated LDMs
- Public Diffusion Models, Private Images: Key-Controlled Inversion for Conditional Reconstruction
- VCT: A Verifiable Transcript System for LLM Conversations
- Have You Ever Seen Them? Entity-level Membership Inference through Interrogating Large Language Models
- Who Owns the AI Recommendation? A Multi-Industry Empirical Map of Brand Category Ownership Across Large Language Models
- Cognitive Digital Twins: Ethical Risks and Governance for AI Systems That Model the Mind
- Managing Procedural Memory in LLM Agents: Control, Adaptation, and Evaluation
- Unmasking LAION-5B: Age, Gender, Race, and Emotion Biases in Large-Scale Image Datasets
- TooBad: Backdoor Diffusion Models with Ultra-Low Poison Rate and Imperceptible Trigger
- HyperQuant: A Rate-Distortion-Optimal Quantization Pipeline for Large Language and Diffusion Models
- Detecting Malicious Agent Skills in the Wild using Attention
- AOHP: An Open-Source OS-Level Agent Harness for Personalized, Efficient and Secure Interaction
- A Watermark for Vision-Language-Action and World Action Models