Key papers

• How Much of the Routing Gap Is Real? Decomposing the Router-to-Oracle Gap into Reproducible Specialist Advantage and Single-Draw Label Noise (2026)
  Teng-Ruei Chen
  4.39
• Piercing Gilbreath's Conjecture: From Deep Number Theory Insights to Fintech and Cybersecurity (2026)
  Vincent Granville
  4.39
• Beyond Refusal: A Same-Lineage Study of Aligned and Abliterated LLMs for Vulnerability Analysis (2026)
  Mingchen Li et al.
  4.39
• DT-Guard: Intent-Driven Reasoning-Active Training for Reasoning-Free LLM Safety Guardrail (2026)
  He Liu et al.
  4.39
• Format Sensitivity Index: Token-Controlled Prompt Wrapper Robustness and Schema Compliance in LLM Benchmarking (2026)
  Deep Pankajbhai Mehta
  4.39
• Faithful, Not Corrective: Message-Format Effects in Multi-Hop Agent Relays Are Tier-Dependent (2026)
  Zayx Shawn
  4.39
• YUKTI: From Natural-Language Situations to Robust, Verifiable Decisions An Uncertainty-Typed Proposition IR, Assumption-Robust Pareto Frontiers, and a Regret Certificate (2026)
  Suyash Mishra
  4.39
• The Verifier is the Curriculum: Execution-Gated Self-Distillation for Cross-Family Game Generation (2026)
  Chenyu Zhou et al.
  4.39
• Scaffolding the Strategist: Architecture-Dependent Reasoning Interventions in Hotelling Spatial Markets (2026)
  Pratyush Singh
  4.39
• A Theory of Least Autonomy in AI (2026)
  Christophe Parisel
  4.39
• Replicating Belief, Not Bits: Epistemic State Replication for Agentic Systems (2026)
  Jun He et al.
  4.39
• LegalFarePlan: A Label-Setting Framework for Fare-Transparent Urban Rail Route Planning under Non-Additive Fare Rules (2026)
  Tanghui Li
  4.39
• From ambiguous utterances to governed reuse classes: canonicalization, quotient invariance, and conditional decidability (2026)
  Cosimo Spera et al.
  4.39
• Minionese: Comprehensive Benchmark and Mechanistic Study of Multilingual LLM Safety (2026)
  Chigozirim Ifebi et al.
  4.39
• When Are Sparse Feature Interventions Actually Localized? Matched Evaluation for SAE-Based Safety Control (2026)
  Daming Luo
  4.39
• Neutralizing Structural Inequality in the Nigerian FinTech Sector (2026)
  Muhammad Abdullahi Said
  4.39
• ANCHOR: Automated Alignment Auditing for CLI Agents on Real-World Harm (2026)
  Kefan Song et al.
  4.39
• Temporary Authority, Permanent Effects: Commit-Time Authorization for LLM Agents (2026)
  Igor Santos-Grueiro
  4.39
• PromptGraph: Graph-Guided Prompt Sanitization for Balancing Privacy and Utility in LLM Inference (2026)
  Chen Gu et al.
  4.39
• Filtering Harmful Actions Isn't Enough: Phantom Transfer in Agentic SDF (2026)
  Chinmayi Dixit
  4.39
• From Checker to Forecaster: Code-Owned Evaluation of Model-Generated Strategic Routes Under Delayed Ground Truth (2026)
  Aleh Manchuliantsau
  4.39
• AdvNav: Behavior-Guided Black-Box Adversarial Attacks on Vision-Language Navigation (2026)
  Chenyang Li et al.
  4.39
• AMT-X: Phase-Structured Multi-Turn Red-Teaming with Checklist-Gated Evaluation (2026)
  Yi Ting Shen et al.
  4.39
• PREF-Gate: Provenance-Constrained Relational Evidence Fusion with Validation-Gated Selection for Graph Fraud Detection (2026)
  Liming Liu et al.
  4.39
• Valid $\ne$ Necessary: Diagnosing Latent Inefficiency in Chain-of-Thought (2026)
  Daeyeop Lee et al.
  4.39
• Mako: A Self-Evolving Agentic Operating System (SE-AOS) for Autonomous Web Exploitation (2026)
  Praneeth Narisetty et al.
  4.39
• Verifier-Guided Twelve-Tone Composition: A Generate-Verify-Repair Harness for Symbolic Music Generation (2026)
  Congren Dai et al.
  4.39
• Closing the Loop: An Access-Control Architecture for Automated, Anomaly-Driven Network Revocation in IoT Deployments (2026)
  Muhammet Emir Korkmaz et al.
  4.39
• Attacking Graph Foundation Models Through Their Shared Representation (2026)
  Pankaj Kumar et al.
  4.39
• SFGA: A Statistics-First Gating Architecture with Adjudicative Escalation for Trustworthy SFT Data Procurement (2026)
  Arther Tian et al.
  4.39
• Learned Interventions in Lean 4 grind (2026)
  Evan Wang et al.
  4.39
• StageGuard: Physiologically Constrained Sleep Staging (2026)
  Juntang Wang et al.
  4.39
• IsabeLLM: Automated Theorem Proving Applied to Formally Verifying Consensus (2026)
  Elliot Jones et al.
  4.33
• Statistical Adversaries: Natural Backdoor-like Features in Vision Datasets (2026)
  Paul K. Mandal et al.
  3.51
• Boosting with List-Decodable Codes (2026)
  Addison Prairie et al.
  3.51
• i-EXAM: Instructable and Explainable Attack Connectivity Graph Modeler (2026)
  Rakesh Podder et al.
  3.51
• Auditing of Unlearning Algorithms (2026)
  Sahasrajit Sarmasarkar et al.
  3.51
• Did We Actually Fix It? An Independent Adversarial Stress-Test of Post-Point-Adjustment Evaluation Metrics for Time-Series Anomaly Detection (2026)
  Zongye Lyu
  3.51
• Language Identification with Succinct Machine-Independent Traces (2026)
  Moses Charikar et al.
  3.51
• Quantum Cryptanalysis on IBM Quantum Hardware: Extending Even--Mansour Period Recovery from $N=4$ to $N=10$ (2026)
  Taebong Kim et al.
  3.51
• Towards an Automated Test of LLM Security Knowledge (2026)
  Shufan Chai et al.
  3.51
• CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization (2026)
  Xinting Liao et al.
  3.51
• Find Before You Fine-Tune: A Diagnostic Study of Small LLMs for Cybersecurity QA (2026)
  Shaswata Mitra et al.
  3.51
• Data Leakage Prevention in Agentic Applications via Preemptive Hardening (2026)
  Akansha Shukla et al.
  3.51
• They'll Verify. They Just Won't Act. How Authority Framing and Laundered Code Turn a Trusted Agentic CI/CD Pipeline Into an Attack Surface (2026)
  Yohann Sidot
  3.51
• Selective Impairment of Motor Recovery from Typing Errors in Parkinson's Disease: A Survival Analysis (2026)
  Navin Bondade
  3.51
• ScoreShield: Differentially Private Release of Similarity Scores (2026)
  Behrooz Razeghi et al.
  3.51
• Many-body Tipping Dynamics of ChatGPT-like AIs (2026)
  Frank Yingjie Huo et al.
  3.51
• Engine-Equal, Human-Unequal: A Reproducible Outcome Skew in Engine-Assessed Equal Chess Positions (2026)
  Jesung Park
  3.51
• Falling Behind Drives Unsafe Development in an Idealised AI Race Experiment (2026)
  Elias Fern\'andez Domingos et al.
  3.51
• NormWorlds-CF: Solver-Verified Counterfactual Normative Reasoning with Metamorphic-Relation GRPO (2026)
  Xinqi Zhang
  2.00
• Agent Step Value: Auditing Evaluator-Channel Reversals in Black-Box Agent Traces (2026)
  Andrew Zhang et al.
  2.00
• Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages (2026)
  Lucas Pinto
  2.00
• Where to Intervene? Benchmarking Fairness-Aware Learning on Differentially Private Synthetic Tabular Data (2026)
  Vin\'icius Gabriel Angelozzi et al.
  2.00
• Reference Traces for Auditing Invisible Weight Updates and Guiding Exact-Budget Protection (2026)
  Zekai Shang
  2.00
• On-Device Deep Research at 4B: Exposure Bounds Faithfulness, Retrieval Bounds Coverage (2026)
  Vinay Kumar Chaganti
  2.00
• Untrusted Authors, Trusted Answers: A Calculus of Fidelity-Graded Translations (2026)
  Christoph Kirsch
  2.00
• Reliability Scales Inversely: Hallucinations Snowball Faster in Bigger Language Models (2026)
  Kushal Chakrabarti
  2.00
• Operational Proto-Introspection in Looped Language Models: Process-Quality Taps, Executable Branching, and the Readout-Control Boundary (2026)
  Jan Kirin
  2.00
• Artificial Epanorthosis: Why large language models overuse a classical rhetorical figure, and how to mitigate it (2026)
  Federico Boggia
  2.00