Infrathon
Claude Opus 5.5 raises the mean KV-cache compression ratio from 6.82% with the human-designed Quant-VideoGen codec to 19.86%.
Benchmarking Language Agents for Real-World, Long-Horizon AI Infra Optimization
AI Research Radar
November 5–11: 15 frontier papers out of 2,333 read.
Frontier
Papers rated as moving the field
How we chooseFrontierPhysics
On FrontierPhysics, no AI configuration among 26 passes more than 46 of 168 end-to-end physics research rollouts.
When Experience Becomes a Backdoor
BadEvolve demonstrates effective task-level backdoor attacks on self-evolving LLM agents across diverse frameworks, benchmarks, and backbone LLMs.
Frontier3DBench
GPT-6.1-sol achieves the highest overall score of 66.02/100 on Frontier3DBench, while all 11 configurations struggle with reference reconstruction.
Mitigating Agentic Deception via Evaluation-Side Partially Observable Games
e-PoG outperforms the strongest baselines by 26.0% in deception mitigation and by up to 39.8% in task success.
Jais 3
Jais 3 matches or outperforms open-weight models of comparable size on Arabic and agentic benchmarks with fewer hallucinations.
Explaining and interrupting collective misalignment in language model systems
Collective Dependence Analysis reduces violations from 93% to 6% in controlled cases using O(N) model evaluations.
Compass 1
Compass 1 achieves a 79.4% overall success rate in open-world embodied navigation, versus 17.8% for the strongest baseline.
Shortcuts and Boundaries
A binary compliance penalty raises safe completion on 32 unseen changing-evidence tasks from 53.1% to 89.1% while both policies achieve full real work.
Meituan-Robotics-0
Meituan-Robotics-0 achieves 93.6% success on RoboTwin 2.0 and outperforms π0.5 on three real-robot tasks.
Authorization Drift in LLM Agents
Frontier-capable LLM agents exhibit authorization drift rates of 0.43 to 0.61 across 13 models and three frameworks.
Multi-step SWE
On the Multi-step SWE benchmark, the strongest of nine frontier models solves only 18.5% of 69 tasks.
A Behavioural Framework for Predicting and Understanding Loss of Control in Frontier Artificial Intelligence Systems
A behavioural framework's loss-of-control score correlates with failure rates in contained long-horizon risk scenarios across 13 frontier models (Pearson r = 0.84).
BEAKER
“Results show that current models can follow coarse experimental procedures in videos, but still struggle to ground actions into operable regions and feasible trajectories, to recognize fine-grained laboratory apparatus, states, and safety norms, and to order steps in mid-length procedures.”
Emergent Biological Capabilities in a Foundation Model for Molecular Interactions
Bonbon shows zero-shot emergent capabilities across functional, residue and bond levels, generalizing to out-of-distribution complexes.
The week by field
Nov 5–11: 281 cleared, 15 frontier
- Security87◆ 2
- Agents44◆ 6
- Safety36◆ 4
- InferenceRising29
- Society23
- Robotics19◆ 2
- 4 moreScience, Reasoning, Training, ML18◆ 1
- Language14
- Vision11
- Security87◆ 2
- 8 moreSociety, Robotics, Language, Vision, Science, Reasoning, Training, ML85◆ 3
- Agents44◆ 6
- Safety36◆ 4
- InferenceRising29
Size: cleared papers◆ and tint: frontier papersRising beyond normal weekly swingInference rising: 4.2% → 10.3% of papers
Models under test
Papers this week that name the model
Labs publishing
Papers this week with an author at the lab
Also cleared
266 papers, by field
Security & privacy85
- TERM: Target-Free Jailbreak Defense via Entropy-Guided Residual ModulationMicrosoft
- MetaPI: Constructing Prompt Injection Benchmarks from LLM Agent BenchmarkNVIDIA
- zkARc: Succinctly Verifiable Proofs for Neurosymbolic Natural Language Formalization and VerificationAmazon
- Scaling Adaptive Prompt Injection Attacks with Agent SwarmsGoogle
- XposedMIA: Exploiting Explainable Artificial Intelligence (XAI) for One-Shot Membership Inference AttacksIBM
All 85 papers in Security & privacy
- MemoryDojo: Benchmarking Poisoning Through LLM Agent MemoryMicrosoft
- Decision Models as Retrieval Gates Against Injection and PollutionGoogle
- Location Matters: Sparse Latent-Variance Attacks for Face Privacy against Diffusion PersonalizationCode
- DiffusedSplitFed: Latent Diffusion-based Obfuscation for Reconstruction-Resilient Split Federated Medical Image SegmentationCode
- BAPLe++: Cross-Model Transferability and Prompt-Level Defense of Backdoors in Medical Foundation ModelsCode
- Beyond Bit Matching: Orthogonal Watermarks for Collusion-Resistant Image FingerprintingCode
- MUTE: Multi-Level Alignment Uncoupling Against Talking-Head Exploitation for Voice ProtectionCode
- RL-ATTACK: A Generalized Black-Box Adversarial Attack on Language Models via Reinforcement LearningCode
- SetPrint: Black-Box LLM Fingerprinting via Setwise Correspondence of Response Changes
- MOAI: Module-Optimizing Architecture for non-interactive secure transformer Inference
- BypassGym: A Benchmark for Evaluating AI Agents' Endpoint Defense Evasion Capabilities
- Domain-Invariant and Language-Model-Assisted Acoustic Side-Channel Attacks on Keyboards
- From Attack Surface to Defense Architecture: Compositional Security for Agentic Coding Harnesses
- One Sequence Is Enough: Fingerprinting Language Models from Fixed-Length Integer Sequences
- MammothDP: Private Boosted Decision Trees Ready for Trusted Hardware and Leakage-free Hyperparameters
- SecureGARD: Dynamic Ontology-Guided Governance for Secure Tool-Using LLM Agents
- Toward Safety-Aware AI Agents in Realistic Social Engineering Attack Scenarios
- Governed Self-Improvement for a Local-First Security Analyzer: Baseline Measurements and a Preregistered Evaluation Protocol for Tantular Cyber
- Bayesian Low-Rank Posteriors for Scalable Membership Inference
- PrivMark: Pixel-Space Steering against Privacy Leakage in Vision-Language Models
- EgoVeil: Personalized Privacy Reasoning for Egocentric Video
- MGE: A Membership Guided Data Extraction Attack against Diffusion Models
- CyberBinGym: Can AI Agents Turn Binary Patches into Real Exploits?
- Mirage: Provable Defense Against LLM KV Cache Timing Side Channels with Topology-Aware Isolation
- The Devil Wears Markdown: Detecting Malicious Intent Hidden in AI Agent Skill Instructions through Hybrid Static, Dynamic, and Agentic Analysis
- Agents of Chaos: A Systematization of Knowledge (SoK) on Malicious Attacks on Agentic AI Systems
- Validating AI-Discovered Cloud Threats on Live Infrastructure: A Multi-Model, Multi-Cloud Study
- Beyond AUROC: Deployment-Aware Evaluation of Model-Internal Jailbreak Detectors
- Tracing MCP Security Vulnerabilities: Constructing and Analyzing an MCP Security Vulnerability Dataset
- A Layer-wise Study of How Style Rewriting Breaks the Calibration of Zero-Shot LLM Fake-News Detectors
- Chatter Attack: Resource Consumption Attack for Large Language Models
- Least-Privilege Auditing of Agent Tool Calls with Intent Graphs and Reasoning-Guided Escalation
- SkillCODER: Towards Auditing and Attribution of Copyright Infringement in LLM Agent Skills
- RoCoMark: Robust Code Watermarking Against LLM-Based Functionality-Preserving Rewriting Attacks
- When Coding Agents Get Arms: Safety Risks Across the Digital-Physical Boundary
- Coding Agents Propagate Vulnerabilities in Code Bases
- On the Relative Security of Prompt Injection: Mechanised Definitions, Defences and Verdicts for LLM Agents
- Data Manipulation Attacks Against Machine Learning Integrated Cyber-Physical Systems
- PRISM: Causal Counterfactual Minimization of Agent Tool Privileges via Information Bottleneck Role Induction
- Treating the Model as an Untrusted Component: Deterministic Policy Enforcement, Human Approval and Taint Tracking Against Prompt Injection in Tool-Using Agents, with Validation on Open-Weights Models
- ShadowBench-MCP: Measuring Shadow Invocations in MCP-Style Tool Routing
- Ultrasound-Assisted Biometric Watermarking Against Face Video Deepfakes
- VeilTune: Secure Fine-Tuning of Private LoRA Adapters on Public Pretrained Language Models
- FALC: An Artifact-Centric Life Cycle Model and Research Agenda for AI Forensics
- StealthyPoison: Manipulating LLM Agent Behavior through Procedural Poisoning of Third-Party Skills
- SNAPRAG: Cost-Efficient Knowledge Extraction Attack on RAG Systems via Adversarial Perturbations
- Detecting Shadow AI: Unsanctioned LLM Agents on Enterprise Endpoints: A Taxonomy, an Open Detector and a Controlled Evaluation
- Say Shibboleth: Fingerprinting the LLMs and Harnesses Behind Agentic Attacks
- Suppression Is Not Erasure: Mechanistic Analysis and Reactivation Boundaries of Subspace Projection for LoRA Backdoor Defense
- I Don't Care How You Hide It, I Care What You Do: Runtime Detection of Malicious ML Models
- AttentionSwitch: Jailbreak Defense for LLMs via Joint Learning of Sparse Attention-Head Selection and Scaling
- Debate to Deceive: Poisoning LLM's Fact-Checking via Game-Theoretic Multi-Agent Debate
- Conflux: Preventing Privilege Escalation with Principal Contexts and System-Level Evaluation
- I Think, Therefore I Attack: Evaluating Thought Injection Attacks on Reasoning Agents
- Notwithstanding Anything to the Contrary, Ignore All Instructions: Measuring Prompt Injection Guards on Legal Contract Review Attacks
- Reconstructing LLM Lineage Trees via Fine-Tuning Path Inference
- NotaryMark: Multi-Bit Watermarking for Black-Box DatasetUsage Auditing in Text-to-Image Diffusion Models
- GreyZone: Production-quality Agentic Application Threats
- AUTEUR: Acoustic Plan Learning for Red-Teaming Text-to-Audio Models
- Logit Laundering: Evading Data-Use Auditing in Large Language Models
- ReplayAudit: Auditing Outsourced LLM Fine-Tuning via Sealed Records and Selective Replay
- GuardrailAgent: An Agentic Framework for Jailbreak Defense
- Interactive Backdoor Attacks in LLM-Based Proactive Dialogue Systems
- SPILLWAY: Separating Target Admission from Sanitization to Defend Tool-Using Agents against Prompt Injection
- Breaking TraintoDefend via Adaptive Neuron Clustering in Cryptanalytic Model Extraction
- Dynamic Watermarking and Action-Conditioned Latent Dynamics for Secure Vision-Based Controllers
- JailBound: FOL-Guided Jailbreak Evaluation for Exploring LLM Behavioral Boundaries
- Embedding-Space Fingerprinting of LoRA Fine-Tunes: Recovering the Training-Data Domain from the Model Artifact Alone
- A Gingerbread House for LLM Agents: Source-Planned Honeypots That Lure Autonomous Attackers into an Isolated Shadow
- What the Middleman Delivers: Answer Behavior, Tool Permissions, and Billed Usage in LLM API Relays
- An Information Bottleneck for Post-hoc Memorization Mitigation in LLMs
- The Promise of Strategic Agentic Scambaiting
- Escalate or Close: How Fast-Tier Models Handle Held-Out Web Attack Families in Two-Tier Triage
- Timing is Everything: Inferring Temporal Membership in Continually Updated Models
- Enforcing Accountable Disclosure in Confidential AI
- Reconciling Privacy with Provenance for Language Model Watermarks
- CoRTeC: A Differentially Private Release for Synthetic Tabular Data, Decodable by Any Generator
- A Bag of Tricks for Defending Deep Learning-based Face Recognition Pipelines against Backdoor Attacks
- Signed but Steerable: Measuring and Closing the Text-Vector Integrity Gap in Offline Retrieval-Augmented Generation
- Signed, Sealed, Leaked: Trust Beyond Signature for MCP Toolchains
Agents38
- Flex-RSI: Benchmarking Recursive Self-Improvement of Frontier AgentsNVIDIA, IBM
- Agent Architecture Cost-Success Trade-offs: A Controlled Evaluation across Enterprise-Style SOPsAmazon
- ReTrace: Distilling Tool-Using Agents into Planner-Free WorkflowsAmazon
- The Memory Trap: Language Agents Overgeneralize from Benign InteractionsAnthropic
- From Handcrafted to Generative: RLVR-Optimized Icon Generation for Alexa+Amazon
All 38 papers in Agents
- Auditable Human-Agent Teams for Autonomous X-ray Experimentation
- From Questions to Decisions: Benchmarking How LLM Agents Acquire Decision-Relevant Information from Human Beings
- PegasusAI Studio: Legible Cyberinfrastructure and Accountable AI Agents for Scientific Workflows
- DEPUTYBENCH: Evaluating Contextual Safety in Enterprise Agents
- Synapse-Control: Making Agent Experience Actionable with Lightweight Runtime Control
- BioXArena-Pro: a research-grade benchmark for autonomous biomedical machine learning agents
- What a CAGE 4 Score Measures: An Audit of a Learned Cyber-Defense Benchmark, Its Reference Defenders, and the Tables Built on Them
- Verifiable Agentic Traffic Class: A Policed Quality-of-Service Mechanism for Agentic AI Traffic, Evaluated from Simulation to Real-Kernel Adversarial Testing
- VIVID: Bringing Tool-Augmented Agents into Enterprises
- Done Is Not Correct: Measuring Silent Failures and Self-Verification Calibration When LLM Agents Take CAD Actions
- Safe by Accident: Crash-Testing LLM-Written Code
- DeepFinance: Can AI Diligence a $100MM Deal?
- GARP: Real-Time Generative Agents on Local Hardware in a Game Engine
- LoMASA: A Local Multi-Agent Sales Assistant with Confidence-Gated Delegation for B2B Quoting
- ScrapeBench: Evaluating Legal Compliance of AI Agents in Website Scraping
- MIRIX-2.0: Multi-Agent Memory System
- Aegis OS: A Transactional Multi-Tenant Operating System Runtime for Eliminating Authorization Fatigue in Autonomous Coding Agents
- Decision Context: Persistent Behavioral Learning for AI Agents
- Enabling Privacy-Preserving Natural Language Access to Industrial Smart Meter Data
- A Multi-Modal Robotic Framework for Physical GUI Interaction Testing
- KG-AST for Cheaper CI/CD: Budgeted Test Selection and AI Bug Review
- The Last Mile to the Real World: Agent-Native Primitives as Reusable Building Blocks for Solving Real-World Tasks
- Signed but Not Safe: Requirement Integrity in Multi-Agent LLM Handoffs
- DelegateEval: Evaluating Platforms for Personal External Service Agents
- Don't Prompt Your Agents. Equip Them: Skillware as a Modular Standard for Sovereign, Auditable, and Deterministic Capability Execution for Logical Systems
- Accepted but Wrong: Silent Failures of LLM Code Repairs after API Migrations
- Benchmarks as Software: Terminal-Bench and the Case for Continuous Benchmarks
- Agentic Behavior Engineering: A Design-by-Contract Methodology for Specifying, Verifying, and Governing Enterprise AI Agents
- Anduin: A sybil-resistant agentic marketplace reputation system
- ViBench 2.0: A Benchmark for Long-Horizon Software Building
- The Harness Is the Model: Agent Scaffolding Changes a Fixed LLM's Compliance with Banking Conduct Rules, Unevenly Across Models
- LifeVal: Evaluating AI Model Performance on Personalized Agentic Consumer Tasks
- Visual Token Routing: Learning to keep tokens that matter
Safety & alignment32
- Double Blind Evals: Resolving the Dual Confidentiality Dilemma in AI Safety AuditingGoogle
- Designing Effective Monitor-Based Interventions for Mitigating Reward Hacking During RLGoogle
- Enmo Safety: External, Non-invasive, Model-agnostic Oversight for Off-the-Shelf Language ModelsCode
- AI Governance Should Prioritize Control and Knowledge Boundaries Over Limiting Intelligence
- Wrong the Same Way: Cross-Prompt Sharing Both Suppresses and Amplifies Reward Hacking in RLVR
All 32 papers in Safety & alignment
- Transparency Disclosures for Chain-of-Thought Monitorability
- Auditing the Judge: Internal Signals for LLM Reasoning Assurance
- Personalized Alignment of Language Models with Human Value Profiles
- A Model Organism for Studying Unverbalized Evaluation Awareness
- META-IR: Meta-Introspective Reasoning for Adaptive Safety Alignment in LLM
- Restoring General and Improving Domain-Specific Safety in Fine-Tuned LLMs Without Retraining
- Coding or Cheating? An Investigation of Frontier Coding LLMs
- What Do Collusion Probes Read? Three Pitfalls in Evaluating Multi-Agent Collusion Monitors
- Hallucination-Guided Unlearning: Using Hallucination Traces to Reveal Overfitted Memories
- R-BED: A Consequence-Aware Framework for System-Level AI Evaluation and Continuous Assurance
- When Safety Becomes An Outlier: Understanding the Retention of LLM Safety Behaviors
- System Prompts Affect Response Bias More Than Discriminability in LLM Safety
- ErosionBench: Probing Alignment Decay in Enterprise Multi-Agent Systems
- Safeguarding Agentic AI in Radiology Workflows: A Multimodel Evaluation of Rule-based and Large Language Model-based Verification
- FraudBench: A Legal Evaluation of AI Deception
- Can the Confidence and Decisions of System 1 Models Be Trusted? An Explainability and Perturbation Audit of Jev-Style Decision Models
- Forcing a Letter Suppresses Refusal: Mechanistic Evidence from WMDP
- Guardrail Erosion in Domain Fine-Tuned Large Language Models: A Paired Audit of Knowledge-Centric Abstention in Biomedical LLMs
- Aligning, Fast and Slow: A Control-Theoretic Oversight Framework for Long-Horizon Trajectories and Maintained Deployments
- BMMS: Policy-Conditioned Measurement of Behavioral Compliance in Language Models
- Goodhart in the Planner: Sampling-Based Control Exploits Rewards Learned from Human Preferences
- Compositional Unalignment: Auditing Safety in Large Vision-Language Models
- Provable Zeroth-Order Safe Exploration for Reinforcement Learning from Human Feedback
- Stress Testing Chain-of-Thought Monitoring Against Covert Misalignment
- SafeStart: Mitigating the Overlooked Safety Cost of Active Expert Reduction in MoE LLM Serving
- Does Context Compression Preserve Refusal Alignment?
- Addition by Subtraction: Toward Mitigating Fragile Safety Alignment in Large Language Models
Inference & efficiency29
- BitDecoding-FP4: Native FP4 KV Cache Decoding via Algorithm-System Co-DesignTencentCode
- Decode Token Parallelism: Memory-Efficient Distributed Attention Parallelism for Scalable LLM DecodingNVIDIA
- Placing LLM Lookup Tables by Per-Layer Prefetch DeadlinesMicrosoft
- ForeSpark: Joint Draft and Verification Budget Scheduling for Parallel Speculative DecodingMeta
- AutoEMatic: Runtime-Adaptive Mixed-Precision LLM Inference via Automatic Exponent-Mantissa RefinementNVIDIA
All 29 papers in Inference & efficiency
- Multi-GPU MoE Inference with Adaptive Expert OffloadingMicrosoft
- PACE: Keeping the Right Tokens in the KV Cache Is Not Enough for Long-Horizon LLM AgentsAmazon
- BridgeKV: Large-Model Prefill and Small-Model Decoding for LLM Inference via KV Cache TranslationMicrosoft
- Capital Efficiency, Isolation, Risk, and Ruin in Generative AI Inference Serving InfrastructureMicrosoft
- FastAFD: When Does Attention-FFN Disaggregation Pay on Rack-Scale NVLink?Code
- FlashOffloading: Low-Latency KV Cache Offloading for Long-Context LLM ServingCode
- Swanada: Split-Encoder Capability-Routed Conformer CTC and Sub-30 MB On-Device Speech Recognition Across Twelve Scripts
- Can KV Cache Be Reused Across Model Updates? A DeepSeek-V4-Flash Case Study
- Ensemble Routing for Prompt-Adaptive Cost-Aware LLM Inference
- Firepool: Brokering Warm GPU Capacity for Agentic and RL Workloads at LinkedIn
- Think or Remember: The Exchange Rate of Agentic Inference
- AutoQS: Automated Quantization Strategy Design with Large Language Models
- Translating KV Caches Between Hybrid-Attention Language Models: MLA Latents, Linear-Attention States and Sparse-Attention Indexers
- System Optimizations for Cost-Efficient Agent Harnesses
- Prune, Don't Merge: Expert Compression of Mixture-of-Experts LLMs on Multilingual Production Tasks
- MNN-Pocket-0.3B Technical Report
- Prediction with Foundation Models under Deployment Budgets
- Inference Validation for Regulated AI: A Governed Decision Pipeline for LLM Serving
- CascadeRec: Deployed Adaptive LLM Compute Allocation for Multi-Tenant Content Recommendation
- AlterNet: Adaptive-Depth Mixture-of-Experts with Learned Halting for Elastic Inference
- AutoCompress: Automated Hardware-Guided Model Compression
- VAR-Q: Tuning-free KV Cache Quantization for Visual Autoregressive Image and Video Generation
- KVGuard: A KV Cache Safety Checker for LLM Serving Systems
- ServerBench: How can your coding agents design specialized LLM serving systems?
AI & society23
- EJAI: Judgment-Preserving Civic AI for Data Center Environmental Justice
- Framing-Induced Vulnerability in Large Language Models: Evidence from Chinese News Headlines
- FAIDER: Fairness-Aware Item-Level Decoding for Large Language Model Recommenders
- Policy Over Purpose: Auditing LLM Content Moderation of Reclaimed Slur Usage
- Too Big to Inspect? Ethically Auditing Large Datasets Through Small Samples
All 23 papers in AI & society
- Behavioral Evaluation of Large Language Models Under Cognitive Biases in HAZOP Studies
- Measuring Generative AI Content in Political Campaigns: Evidence of Rising Adoption and Non-Disclosure
- State Media at Retrieval Time: A Large-Scale Audit of Russian State-Aligned Sources in AI Chatbot Responses
- CHO-Bench: How Do Context Hints Affect Ad Rankings in LLM Advertising?
- Detecting AI-Authored Text in Authentic Student Writing: Evidence from PISA 2025
- Fairness in Explainable AI Narratives: Decomposing Social Disparities in LLM-Generated Explanations
- Curbside Chatbots: Auditing the Mental Health Interventions Large Language Models Recommend to Clinicians, Caregivers, and Patients
- Same Facts, Different Remedy: Linguistic Profiling and Arbitrariness in LLM and System One Decisions
- What to Delegate, What to Preserve: Human Preferences for Agentic AI Across Everyday Life and Work
- Strategic Competition and Convergence in LLM Leaderboards
- Reweighting is not Harmless: Understanding the Role of Dataset Characteristics
- Adversarial Multiplicity: Cherry-Picked Models and the Limits of Disclosure
- The Popular Get More Popular: Auditing Brand Bias in Generative Search
- Responsible GenAI Use During Active Armed Conflict: Evidence from Sudan's Emergency Response Rooms
- Who Can Evaluate AI Independently? A Framework for Analysing Evaluation Infrastructure
- Just the Facts: How Dialogues with AI Reduce Conspiracy Beliefs
- Does Frontier AI Strengthen Cyber Defense? Evidence from Project Glasswing
- Beyond accuracy: what AI assurance needs to establish in public services
Robotics & embodied AI17
- HARP-Bench: Benchmarking Robot Manipulation Around Active HumansMetaCode
- Interactive World Simulation with Cosmos for Physical AINVIDIACode
- DeltaForge: A Post-hoc Mixture-of-Experts VLA for Multi-Domain Autonomous DrivingNVIDIA
- Open-HumanWAM: An Open Recipe for Scaling Human and Robot Experience for Embodied Foundation ModelsNVIDIA
- MotoVerse: A Benchmark for Predictive Perception of Motorized Two-Wheelers in Dense Unstructured TrafficCode
All 17 papers in Robotics & embodied AI
- From Depth to Safe Depth: Uncertainty-Aware Monocular Perception for Mobile Robot Navigation
- Camera equalisation of an 11.5x drop in environmental illuminance: four imitation-learning policies on an SO-101 arm
- LEASH: Stay Safe in Agentic Robotics
- From Game Generation to Game Playing: Fully Automated Ability Training for Vision-Language-Action Models
- Breaking survivorship bias in driving world models
- DriveFoundry: Co-Evolving Data and Model for Efficient Closed-Loop Iteration in Autonomous Driving
- Planners Exploit Deterministic Error: Dithered Low-Bit World Models
- Probably Safe Sim-to-Real Transfer with Lifetime Validity
- Can LLMs Debug Robots by Editing Their Weights?
- Spatial-UMI-Dream: Scaling World Action Models with Metric Geometry for UMI Manipulation
- RoboARCade: Benchmarking Agentic Robot Control
- Jev with Foresight: Safe Language-Guided Autonomous Driving via a Language World Model
Language & speech14
- Yuanbao-Reranker: Document-Grounded Synthesis and Preference Distillation for Text RerankingTencent
- The Indic Token Tax in 2026: Measuring Tokenizer. Disparity and Its Cost for the Languages of IndiaCode
- A Multi-Word Tokenizer for 204 Languages That Needs Fewer Tokens Than GPT-4o's, Including on English
- Inherited Weights: Measuring the Tokenization Premium and the Sovereignty Cost of Language Models for the Global South
- IFRS Compliance Verification: An Open Multi-LLM Benchmark and Content-Conditioned Ensemble Classifier
All 14 papers in Language & speech
- Walrus Technical Report
- PolyRoute: A Twelve-Language and Code-Mixed Benchmark for Typed-Decision Routing, with Baselines and a Leaderboard
- Alania-2: Turkish text-to-speech with instant and fine-tuned voice cloning
- Earlybird: Predictive Streaming Speech Recognition in 29M Parameters
- Retrofitting Unbounded Context onto Stock Language Models
- M5-TTS: A Mini-Scale Multilingual and Multi-Dialect Text-to-Speech Language Model with Multi-Token Prediction
- LAION-Audio-XL: A Web-Scale Corpus and Scaling Studies for Contrastive Language-Audio Pretraining
- ZeroRAG: Adaptive Semantic Peer Routing for Decentralized Retrieval-Augmented Generation
- How Well Can Large Language Models Impersonate Individual Authors? A Comparative Evaluation
Vision & multimodal11
- Beyond Real or Fake: A Dual-Channel Authenticity and Reasoning Protocol for Photographic AssessmentGoogle
- Video-Coder: Reinforcing Agentic Coding for Video GenerationMicrosoft
- WeLM-Image: A Unified Image Generation Model towards Visual IntelligenceTencent
- Seeing Is Not Believing: Counterfactual Lesion Erasure Teaches a Small Medical Vision-Language Agent to Report Only What the Image Shows
- Towards Reliable VLM Judges: State-Conditional Invariance and Presentation-Aware Diagnostics
All 11 papers in Vision & multimodal
- ScreenProof: Where Is This Screen Wrong? Finding UI Faults Without a Reference
- From Weapons to Events: Evaluating Evidential Support in Multimodal AI for Policing
- Direct Trajectory Optimization for Inference-Time Alignment of Text-to-Image Diffusion Models
- InnerLogic-Surg: An Open Surgical Foundation Model via Continual Pretraining
- When GeoFMs Meet VHR: Benchmarking Foundation Models on Very High Resolution Earth Observation Imagery
- Trick-or-Treat: Adversarial Evaluation of Multimodal Medical Reasoning with Synthetic Images
Science & health7
- HyPLM Technical ReportCode
- An evidence-traceable benchmark of clinical reasoning in large language models
- Ninety-Six Ways to Measure a Scribe: Fragmented Evaluation of Ambient Scribes in Healthcare
- FAIR Universe Weak Lensing ML Uncertainty Challenge: Dataset and Cosmological Inference Competition
- TrialDesignBench: Benchmarking Language Models for Phase-I Trial Design Against Clinical Knowledge