Skip to content
Oct 13 · Tuesday

All papers

Found 1032 papers
Filter6

Attack type

Jailbreaks76Prompt injection215Poisoning and backdoors96Model tampering20Extraction and theft36Detection evasion33Denial of service5Malicious use18

Attack technique

Gradient and representation optimization21Automated search52Multi-turn escalation8Encoding and obfuscation16Role-play and persuasion3Conversation structure manipulation3Reasoning-chain manipulation4Cross-modal payloads9Defense-targeted26Dormant triggers35Selection manipulation11

Model risks

Deception and scheming58Reward hacking36Misalignment and value divergence40Dangerous agent actions79Dangerous capabilities25Inappropriate refusal54Hallucination and fabrication33Bias and discrimination20

Defense type

Model hardening28Inference-time intervention46Detection and filtering49Instruction–data separation51Permissions and approval69Monitoring and auditing60Watermarking and provenance18

Impact

Harmful content104Unauthorized actions211Data and privacy leaks82Output manipulation174Availability and reliability56Oversight failure83Increased misuse capability15

System type

Models and APIs410LLM applications132AI Agent622coding agent127science agent30web agent34Browser and computer use19Multi-agent systems82Embodied systems and robotics27

Attacker access

Black-box34Gray-box6White-box24No-box (offline transfer)3

Attack surface / entry point

User input165Web pages and interfaces62Documents and messages70Tool responses102Tool and skill descriptions63Code repositories and dependencies27Retrieval corpus13Memory writes41Inter-agent messages28Images, audio and video32Training data12Weights and gradients34Distribution channels50Inference services and hardware7

Layer

Prompt layer253Model layer425Application layer607Training and data layer239Infrastructure layer22

System component

MCP and skills82Memory78RAG9Guardrails and judges58Monitors79Reasoning chains44Permissions and sandboxing90Fine-tuning and open weights14Vision54Audio18Video5Image generation20Embeddings4Multimodal90
Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 227papers have no tags yet and are not included in filtered results.
System type：Models and APIsSystem type：AI AgentSystem type：Multi-agent systemsSystem type：Embodied systems and roboticsLayer：Model layerLayer：Application layerClear all

Another 227papers do not have category tags yet and are not included in filtered results.

  1. AttacksarXiv 2610.13928

    AI models can provably hide arbitrary capabilities

    Proposes digital lock compilation, encrypting executable networks into host weights to hide arbitrary capabilities

    arXiv
    2610.13928
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Gemma 3 270M
    AbstractThe authors use a digital lock to hide an executable capability and state that elicitation on the chemistry and parity tasks did not recover the capability.

    The authors construct a class of backdoors that encrypt an executable network into the weights, used to discuss the limits of white-box elicitation audits under supply chain attacks.

    Paper analysisRead analysis
  2. OtherarXiv 2610.13912

    RSI-Anything: Unlocking Recursive Self-Improvement with World Models

    Proposes RSI-Anything, which uses a world model to select candidates and retain policy updates.

    arXiv
    2610.13912
    Published
    Layer
    Application layer
    Setting
    coding agent, Browser and computer use
    Evaluated models
    ReasonFlux-Coder-7B, 4B policy, W1
    AbstractRSI-Anything integrates world models into recursive self-improvement through prediction, feedback, and retention checks, and reports gains across multiple domains.

    RSI-Anything is a cross-domain framework: it uses world model predictions to decide which candidates to take external feedback for, then uses the feedback to decide whether the policy, workflow, or predictor is updated, and uses checks to decide which pair is retained. The difficulty to address is the limited interaction budget, and the fact that once the policy changes, the candidate distribution faced by the predictor also changes. The authors argue that improving only one side does not guarantee that the assembled system will be better.

    Paper analysisRead analysis
  5. Evaluation and benchmarksarXiv 2610.13710

    OpenSocialJax: Evaluating Social Adaptation in Sequential Social Dilemmas with Open-Ended Environment Dynamics

    Evaluating multi-agent adaptation in social dilemmas with hidden rules using OpenSocialJax

    arXiv
    2610.13710
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems
    Evaluated models
    GPT-6-Luna, GPT-6-Sol, Claude Opus 5.5 and others8
    AbstractThe authors report that rule discovery often improves payoffs, but public good cleaning becomes concentrated, and shared resources are eventually depleted.

    OpenSocialJax evaluates whether LLMs can simultaneously adapt to unfamiliar environmental mechanisms and other agents in sequential social dilemmas.

    Paper analysisRead analysis
  7. OtherarXiv 2610.13491

    EviRepair: Evidence-Driven Agents for Reliable Vulnerability Repair

    Proposing EviRepair, an evidence-driven multi-agent framework for repository-level vulnerability repair

    arXiv
    2610.13491
    Published
    Layer
    Application layer
    Setting
    coding agent, Multi-agent systems
    Evaluated models
    DeepSeek-V4.1-Flash, GLM-5.2, Qwen3.8-27B
    AbstractEviRepair uses a reviewed vulnerability model for repository-level repair.

    EviRepair is a repository-level automated vulnerability repair framework for sparse vulnerability reports, using a reviewed vulnerability model to separate diagnosis from patch synthesis.

    Paper analysisRead analysis
  8. OtherarXiv 2610.13479

    A local agentic AI framework generates edge-case scenarios for safety assurance

    Proposes a local dual-model closed loop to generate edge scenarios for autonomous system safety assurance

    arXiv
    2610.13479
    Published
    Layer
    Application layer
    Setting
    LLM applications, Embodied systems and robotics
    Evaluated models
    PPO
    AbstractLocal closed-loop generation of JSON edge cases.

    The authors built a local agentic loop that runs on a laptop to generate edge cases for safety arguments of autonomous systems, rather than releasing a new public test set.

    Paper analysisRead analysis
  10. OtherarXiv 2610.13372

    Toward Safe World Models: Risks, Defenses, and Evaluation

    A survey of risks, defenses, and evaluation in the six-stage lifecycle of world models

    arXiv
    2610.13372
    Published
    Layer
    Application layer
    Setting
    Embodied systems and robotics
    AbstractSafety is written as a property of the prediction-decision closed loop; claims must be linked to observed consequences and versions.

    World model safety is organized by the nature of the prediction-decision loop, not by isolated prediction scores.

    Paper analysisRead analysis
  14. InterpretabilityarXiv 2610.13078

    Attention sees, MLP refuses: tracing the safety pathway from attention to MLP in LLMs

    Trace attention to identify harmfulness to the MLP refusal write pathway and propose Sentinel

    arXiv
    2610.13078
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Llama-3.1-8B-Instruct, Qwen2.5-7B-Instruct, Qwen3-8B-Instruct and others18
    AbstractAttention updates are more readable for harmfulness, MLP updates write refusal, and Sentinel routes alerts using two-path signals.

    The authors reframe jailbreaking and excessive refusal of aligned models as a pathway problem on the residual stream from recognition to writing, and accordingly provide conditional reminders before decoding.

    Paper analysisRead analysis
  15. InterpretabilityarXiv 2610.12989

    GPT-2-Small, Gemma-2-2B, and Llama-3.1-8B have interpretable narrative mechanisms

    Using NVC to locate and ablate narrative-related components in language models

    arXiv
    2610.12989
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    GPT-2-Small-124M, Gemma-2-2B, Llama-3.1-8B
    AbstractNeurons and SAE features related to narrativity across three models, when ablated, changed some narrative-related behaviors.

    The authors connect a three-dimensional definition from narratology to mechanistic interventions, examining whether components related to narrativity in pretrained language models participate in prediction, generation, and downstream tasks.

    Paper analysisRead analysis
  16. InterpretabilityarXiv 2610.12842

    Neural networks are more modular than single neurons suggest

    Using a three-level nested decomposition to test whether mixed-selectivity networks can be functionally separated on subspaces

    arXiv
    2610.12842
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Pythia 410M, Pythia 160M, GPT-2 XL and others18
    AbstractUnit non-specialization and subspace separability can coexist.

    The authors propose a nested definition of modularity to separate "neurons not being specialized" from "computation not being separable."

    Paper analysisRead analysis
  17. Risky behaviorarXiv 2610.12822

    Reward hacking as a transmissible strategy in multi-agent coding systems

    Show that reward hacking can propagate through code, memory, and plans in multi-agent coding systems

    arXiv
    2610.12822
    Published
    Layer
    Application layer
    Setting
    Multi-agent systems, coding agent
    Evaluated models
    Qwen2.5-Coder-14B, Gemma-3-12B, Qwen2.5-Coder-7B and others16
    AbstractThe authors state that reward hacking can spread across models and affect routing and aggregation, and that review can reduce delivery.

    Reward hacking induced by coding RL can leave the source agent and enter subsequent collaboration and system coordination.

    Paper analysisRead analysis
  18. Evaluation and benchmarksarXiv 2610.12758

    DistressBench: Evaluating multi-dimensional crisis support in large language models

    Introduces DistressBench, evaluating models' multidimensional support in crisis conversations

    arXiv
    2610.12758
    Published
    Layer
    Model layer
    Setting
    Models and APIs
    Evaluated models
    Muse Spark 1.2, Muse Spark 1.3, Muse Glimmer 30B and others20
    AbstractThe model is more restrained and acts less.

    DistressBench uses SSH dialogues written and revised by clinical personnel, and evaluates how much necessary care general chat models deliver according to weighted atomic criteria, rather than whether they refuse. When users disclose suicidal ideation or self-harm to a chat model, the authors consider the expected behavior to be recognizing the crisis, calibrating empathy, de-escalating, and referring to human help. Refusal-based evaluations cannot detect failing to act after recognizing the crisis.

    Paper analysisRead analysis
  19. Evaluation and benchmarksarXiv 2610.12569

    SWE-Duel: Toward a Self-Expanding Arena for Coding Agents

    Proposes SWE-Duel, which lets coding agents mutually evaluate, create problems, and blindly review fixes during evaluation.

    arXiv
    2610.12569
    Published
    Layer
    Application layer
    Setting
    coding agent
    Evaluated models
    GPT-5.5, Claude Opus 4.8, Kimi K2.7 Code and others11
    AbstractThe mutual evaluation arena creates questions during evaluation.

    SWE-Duel is a coding agent arena that generates tasks and judges only at evaluation time, to avoid leakage and saturation of static benchmarks.

    Paper analysisRead analysis
Load more