PreprintNot peer reviewedArtificial IntelligenceA preprint reports LLMs show stronger illusory pattern perception than humansOriginal title: Illusory Pattern Perception Drives Spurious Inference in Large Language ModelsA preprint finds large language models tend to infer false patterns from random data more readily than human participants do.Peihua Mai et al. · Added 8 Oct 2026
PreprintNot peer reviewedFrom HuaweiArtificial IntelligenceA preprint reports LLMs often answer even when given context is insufficientOriginal title: Rethinking Faithfulness in LLMs: A Pairwise Context-Sensitive PerspectiveA preprint introduces a paired-question benchmark finding most LLMs fabricate answers rather than abstaining when context is insufficient.Added 8 Oct 2026
Journal article✓ Peer reviewedQuantum ComputingModel captures imperfect Rydberg blockade effects in atomic superatomsOriginal title: Imperfect blockade in Rydberg superatomsResearchers derive and test a model describing imperfect Rydberg blockade in atomic ensembles used for quantum network nodes.Valentin Magro et al. · Added 8 Oct 2026
PreprintNot peer reviewedArtificial IntelligenceResearchers propose communication-efficient federated reasoning for medical LLMsOriginal title: FedCoT: Communication-Efficient Federated Reasoning Enhancement for Large Language ModelsA preprint describes FedCoT, a federated learning framework that improves LLM reasoning on medical QA while cutting communication overhead and keeping data local.Chuan Li et al. · Added 8 Oct 2026
PreprintNot peer reviewedFrom summary.aiArtificial IntelligenceA preprint proposes a semantic-aware distillation method for shrinking large language modelsOriginal title: WASD: Wasserstein-based Knowledge Distillation for Large Language ModelsResearchers report that aligning teacher and student models by token meaning, not just probability, improves knowledge distillation across several LLM families.Added 8 Oct 2026
PreprintNot peer reviewedQuantum ComputingA preprint describes an optimised silicon qubit well design to boost valley splittingOriginal title: Optimization of Si/SiGe Heterostructures for Large and Robust Valley Splitting in Silicon QubitsResearchers propose a computationally optimised Si/SiGe profile, the modulated wiggle well, to enhance and stabilise valley splitting in silicon qubits.Abel Thayil et al. · Added 8 Oct 2026
PreprintNot peer reviewedFrom Microsoft ResearchArtificial IntelligencePreprint proposes benchmark and two world-model approaches for language-guided navigationOriginal title: Language-Conditioned World Modeling for Visual NavigationA preprint introduces a large instruction-following navigation benchmark and compares two world-model paradigms for guiding robots by language alone.Added 8 Oct 2026
Journal article✓ Peer reviewedFrom NVIDIAQuantum ComputingNew unitary circuit preserves surface-code distance while preparing logical statesOriginal title: Unitary fault-tolerant encoding of Pauli states in surface codesResearchers present a local, unitary encoding scheme for surface codes that keeps full error protection as code size grows.Added 8 Oct 2026 · Released Jan 2026
PreprintNot peer reviewedFrom XiaohongshuArtificial IntelligenceA preprint proposes unbiased reward models trained from implicit LLM feedback signalsOriginal title: Unbiased Reward Modeling from Implicit Feedback for LLM AlignmentA preprint describes ImplicitRM, a method for building reward models from passive signals like clicks and copies rather than explicit ratings.Added 8 Oct 2026
PreprintNot peer reviewedFrom PleIAsArtificial IntelligenceResearchers release Common Corpus, a two trillion token open dataset for LLM trainingOriginal title: Common Corpus: The Largest Collection of Ethical Data for LLM Pre-TrainingA preprint presents Common Corpus, a large openly licensed dataset for training language models without copyright concerns.Added 8 Oct 2026
PreprintNot peer reviewedArtificial IntelligenceResearchers propose SEER framework to shorten AI chain-of-thought in coding tasksOriginal title: Reasoning Efficiently Through Adaptive Chain-of-Thought Compression: A Self-Optimizing FrameworkA preprint reports that trimming AI reasoning length can cut truncation and loops while keeping coding accuracy.Kerui Huang et al. · Added 8 Oct 2026 · Released Mar 2026
PreprintNot peer reviewedArtificial IntelligenceResearchers report diffusion language models beat autoregressive ones on constraint-heavy puzzlesOriginal title: Blackboard Intelligence Can Surpass Autoregressive on Globally Constrained ProblemsA preprint finds that letting a diffusion language model revise a whole solution canvas, guided by its own confidence, outperforms token-by-token generation on constrained puzzles.Woosang Jeon et al. · Added 8 Oct 2026
PreprintNot peer reviewedArtificial IntelligenceA preprint proposes hash retrieval method to speed up long-context LLM inferenceOriginal title: HHR: Hierarchical Hash Retrieval for Efficient LLM GenerationResearchers report a two-stage hashing approach that improves retrieval accuracy and speeds up long-context LLM decoding.Lianjun Liu et al. · Added 8 Oct 2026
PreprintNot peer reviewedFrom NVIDIAArtificial IntelligenceResearchers report self-reflection training cuts prompt-injection success in LLM agentsOriginal title: Self-Reflection Fine-Tuning: Enhancing Agent Security against Prompt Injection Attacks from Failure ExperienceA preprint describes a fine-tuning method that teaches LLM agents to learn from failed attacks, sharply lowering injection success rates.Added 8 Oct 2026
PreprintNot peer reviewedFrom Fidelity InvestmentsArtificial IntelligenceA preprint reports a method to stop agent tasks from corrupting shared memoryOriginal title: Safe Context Switching for Agents in the Wild: Mitigating Subspace Interference via Orthogonal AdaptationResearchers propose AURA, a geometric method to stop sequential agent tasks from contaminating each other's residual activations.Added 8 Oct 2026
PreprintNot peer reviewedArtificial IntelligenceResearchers report visual tool-use harness boosts multi-image reasoning in MLLMsOriginal title: Rethinking Multi-Image Re-Representation in Multi-Image UnderstandingA preprint reports that letting models build visual intermediates, not just text, helps most on tasks needing precise visual evidence.Gengyuan Zhang et al. · Added 8 Oct 2026
PreprintNot peer reviewedArtificial IntelligencePreprint reports a training-free sentence check for unsupported RAG contentOriginal title: Sentence-Level Context Sensitivity as a Training-Free Detector of Unsupported Content, Evaluated Against Trained VerifiersA preprint evaluates a cost-free likelihood-contrast signal for spotting unsupported sentences in RAG outputs against trained verifiers.Mohamed Aly Bouke · Added 8 Oct 2026
PreprintNot peer reviewedFrom FujitsuQuantum ComputingPreprint proposes Clifford-only fidelity estimation for quantum resource statesOriginal title: Efficient Fidelity Estimation of General Quantum Resource States via Clifford CircuitsA preprint reports a Clifford-gate protocol that estimates resource-state fidelity with fewer samples than standard methods.Added 8 Oct 2026
PreprintNot peer reviewedArtificial IntelligencePreprint describes small model trained to manage its own long-context retrievalOriginal title: Trained Agentic Context ManagementA preprint reports that a fine-tuned small model with a tiny context window can match a frontier model's long-document performance.Bryce Sandlund · Added 8 Oct 2026
PreprintNot peer reviewedQuantum ComputingPreprint finds conditions for transversal Clifford gates on self-dual CSS codesOriginal title: Clifford gates with logical transversality for self-dual CSS codesA preprint derives when high-rate self-dual CSS codes allow logical Hadamard and phase gates via transversal physical gates.Theerapat Tansuwannont et al. · Added 8 Oct 2026
