Every paper checked — nothing hidden
All papers
Every paper checked in the week of 12–18 Oct 2026 · AI safety and reliability in deployment — nothing is hidden. Brief papers first.
All3,579In brief3More (on-topic)755Not relevant2,801Not assessed3Older63Could not confirm4Blogs with no paper8All blogs8
| Paper | Topic | Relevance | Feedback |
|---|---|---|---|
| Practical wavelength selection for quantum-augmented networks over shared optical fiber Develops a wavelength selection framework for running quantum-augmented error-corrected networks over optical fiber shared with classical traffic. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Addresses physical quantum optical networking, entirely unrelated to AI deployment safety. | |
| Blind Watermarking of Software Repositories via Module Placement Introduces a software watermarking technique that embeds identifiers by altering module placement rather than modifying source text. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Focuses on repository module watermarking for traditional software provenance, not AI safety. | |
| Efficient Agnostic Learning of Univariate Densities Under Local Differential Privacy Proves theoretical sample bounds and presents algorithms for learning univariate probability distributions under local differential privacy. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Theoretical density estimation under differential privacy, unrelated to operational AI deployment safety. | |
| NOMAI: Scalable Federated Learning with Private Weights Presents a secure multiparty computation system for federated learning that protects model weights, user data, and system integrity at scale. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Focuses on multi-party computation protocols for federated learning rather than AI safety or reliability. | |
| Reaction Attack on CKKS Deployments with Asymptotically Optimal Query Complexity Presents a reaction leakage attack that recovers secret keys in homomorphic encryption deployments using client validity responses. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Studies cryptographic reaction attacks against CKKS encryption deployments, not AI safety. | |
| Optimal Reduced Representatives of Quadratic APN Permutations in Odd Dimension Proves existence conditions for optimal reduced polynomial representatives of quadratic almost perfect nonlinear permutations in odd dimensions. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Pure mathematics concerning APN permutations and finite fields, not AI safety. | |
| Amora: Communication-Efficient Private Approximate Nearest Neighbors Search Presents a communication-efficient private approximate nearest neighbors search system for vector databases using partitioned graph indices and private information retrieval. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Addresses cryptographic private nearest neighbor search in vector databases, not AI system reliability. | |
| Explicit reduced APN permutations: ramification and optimal Galois-ring lifts Derives explicit reduced almost perfect nonlinear permutations and analyzes their ramification and lifts over Galois rings. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Mathematical work on APN permutations in finite fields, not AI safety. | |
| On the Complexity of Symmetric Boolean Functions in Multiparty Private Simultaneous Messages Analyzes the communication complexity and lower bounds of symmetric Boolean functions in multiparty private simultaneous message protocols. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Deals with theoretical multiparty computation and Boolean function complexity, not AI safety. | |
| Combining LLM and Dynamic Analysis for Universal Testing of Static Application Security Testing Tools Uses large language models and dynamic feedback to generate code variants and test static application security analysis tools. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Evaluates static analysis tools for traditional software rather than addressing AI deployment safety. | |
| Evaluating and Improving Static Application Security Testing Tools for Rust Evaluates static analysis security testing tools on real-world Rust vulnerabilities and presents an improved framework for vulnerability detection. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Focuses on static security testing of Rust programs, not AI safety or deployment risks. | |
| Benchmarking Post-Quantum Key Encapsulation in S3-Compatible Object Storage under Client Concurrency Measures the performance impact of post-quantum key encapsulation algorithms during per-object client-side encryption in concurrent cloud storage systems. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Focuses on post-quantum cryptography in cloud object storage, not AI safety or reliability. | |
| Smoothness of Shortest Isogenies to Frobenius Conjugates Proves the smoothness of minimum degree isogenies to Frobenius conjugates, confirming an attack complexity for supersingular elliptic curves. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Pure mathematical cryptography paper unrelated to AI safety. | |
| NEST: Stale-Robust Tokenized-Stock Measurement for a Crypto-Treasury LETF Stack Proposes a measurement framework to evaluate the informational value of off-hours tokenized stock prices relative to listed market moves. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Financial econometrics for tokenized stocks unrelated to AI safety. | |
| On the Robustness of Gossip Protocols in Modern Blockchains Evaluates network topology robustness and adversary resistance in gossip protocols used by modern committee-based blockchains. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Blockchain gossip network protocol study unrelated to AI safety. | |
| Pricing the Unprovable: Ambiguity, Turnstile Runs, and the Verifiability Discount in Crypto Protocols Develops a framework and empirical model to measure market valuation discounts following unverified vulnerability repairs in privacy-preserving crypto protocols. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Financial crypto-protocol market analysis unrelated to AI deployment safety. | |
| SMT-Boosted Security Types and Correctness for Low-Level MPC Develops a verification type theory using SMT solvers to enforce security and correctness in low-level secure multi-party computation. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Studies cryptography and MPC verification, not AI safety deployment. | |
| MAppPrint: Robust and Real-time Mobile App Fingerprinting via Traffic Behavior Correlation Graphs Introduces a real-time mobile app fingerprinting method based on traffic behavior correlation graphs to identify apps from encrypted network flows. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Network traffic fingerprinting paper unrelated to AI system safety. | |
| Subgraph Counting with Local Node Differential Privacy Develops an algorithm with tight error bounds for counting subgraphs under local node differential privacy constraints. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Theoretical graph privacy paper unrelated to AI deployment safety. | |
| TurboMamba: Faster Inference, Faster Training, and Better Prediction with Lexical Memory Proposes a hybrid architecture combining global state-space modeling with local transformer attention and lexical memory for efficient inference. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Does not address AI safety, jailbreaking, or guardrail systems. | |
| APRiL: An Open-ended Benchmark for Chinese Social Awareness in Large Language Models Presents a benchmark with open-ended tasks and rubrics to assess social awareness and response generation in Chinese. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Focuses on social awareness benchmarking rather than safety or guardrails. | |
| Gibbs-Gap Policy Optimization for Off-Policy RL with Verifiable Rewards Introduces an off-policy reinforcement learning algorithm for language models using verifiable rewards without importance sampling weights. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Focuses on RL training algorithms rather than deployment safety or guardrails. | |
| PEDBench: Can Speech Models Detect Phonetic Errors? Introduces a benchmark combining natural, scripted, and synthetic audio to evaluate phonetic error detection in speech models. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Does not study AI safety, jailbreaking, or deployment guardrails. | |
| Does Morphological Correctness Matter for Tokenization? A Scrambled Control Study in Tagalog Tests whether linguistically accurate morpheme boundaries in tokenizers improve translation performance compared to scrambled control segmentations. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Does not address AI safety, robustness, or reliability guardrails. | |
| CogDrift-4D: Hierarchical Multimodal Modeling of Temporal Behavioral Decay for MCI Detection Proposes a hierarchical multimodal transformer to model long-term behavioral changes across multi-session interactions for cognitive impairment detection. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Does not address AI safety, jailbreaks, or deployment guardrails. | |
| Task-Centric Multimodal Learning for Cognitive Impairment Detection Evaluates a multimodal screening method using voice, video, and text from voice-assistant interactions to detect cognitive impairment. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Does not focus on AI safety, alignment, or guardrails. | |
| VELD-QA: A Large-Scale Dataset for Visual Evidence Grounding and Question Answering in Long Documents Presents a benchmark with human-annotated visual evidence to evaluate retrieval and question answering across long documents. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Does not address deployment safety, jailbreaks, or red-teaming. | |
| One Bias Score, Two Factors: How Mitigations Affect Option Bias in Multiple-Choice Evaluation Deconstructs multiple-choice option bias in language models into two factors tracking answer flips and positional skew. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Focuses on benchmark evaluation artifacts rather than deployment safety guardrails. | |
| Learning When to Think: Utility-Guided Adaptive Reasoning for Efficient Text Retrieval Proposes an adaptive reasoning framework that supervises intermediate retrieval states and stops reasoning steps dynamically to cut latency. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Does not address safety alignment, red teaming, or guardrails. | |
| Compression Does Not End at Selection: Post-Selection Positional Geometry for Sparse KV Caches The paper proposes RoPERelay to reorganize positional geometry and rotary embeddings after sparse KV cache token eviction. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Does not address AI safety risks, alignment, or guardrail systems. | |
| From Observation to Intervention: Distilling Self-Generated In-Context Demonstrations via Executable Causal Intervention The paper introduces a causal intervention framework to distill transferable self-generated in-context learning demonstrations. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Focuses on reasoning distillation rather than deployment safety or jailbreaks. | |
| Stop Training Against Soft Users: Mentalized Adversarial Self-Play for Proactive Dialogue The paper introduces an adversarial self-play framework modeling user mental states to improve proactive dialogue agent planning. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Focuses on dialogue policy robustness rather than AI safety or guardrails. | |
| Automatic Sentence Completion Using Bengali Audio Speech with a Transformer-Based Early Speech Recognition System The paper presents an automated speech recognition and sentence completion system to assist dementia patients speaking Bengali. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Not centered on safety alignment, hallucination prevention, or red teaming. | |
| From Long Context to Selective Memory:A Survey of Memory-Centric Compression for Large Language Models The paper surveys memory-centric compression techniques for managing inference resources and context windows in large language models. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Does not focus on deployment safety, jailbreaks, or guardrails. | |
| Beyond Retrieval Failure: Understanding Why LLMs Miss Evidence in Multi-Hop RAG The paper introduces ExecutionRAG to show how search drift causes apparent retrieval failures in multi-hop question answering. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Analyzes retrieval coverage rather than safety, jailbreaks, or malicious behavior. | |
| KBS: A Korean Benchmark Suite for Foundation Model Evaluation The paper introduces the Korean Benchmark Suite to evaluate foundation models across six natural language capabilities. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Focuses on broad evaluation capabilities rather than safety or reliability guardrails. | |
| What Should We Remove? Evaluating the Reliability and Costs of RLVR Data Decontamination The paper audits data decontamination methods on math corpora to quantify false removals and benchmark leakage risks. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Studies data overlap rather than jailbreaks, guardrails, or deployment safety. | |
| FGSS-BO: Feedback-Guided Hypothesis Management for LLM-Based Bayesian Optimization The paper proposes FGSS-BO, adding structured hypothesis management to language-model-guided Bayesian optimization. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Does not investigate deployment safety, hallucinations, or alignment vulnerabilities. | |
| LLM-Based Weak Annotation for Cybersecurity Tweet Classification: A Teacher-Student Framework The paper presents a teacher-student framework using language models for weak annotation to classify cybersecurity tweets. Peer-reviewed | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Addresses cybersecurity text classification, not model safety, guardrails, or jailbreaks. | |
| TimeCap: A Scalable Open-Source Pipeline for Captioning Wearable Health Time-Series Data The paper introduces TimeCap, an open deterministic pipeline for generating signal-grounded captions from wearable health time-series data. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Does not address safety alignment, red teaming, jailbreaks, or guardrails. | |
| PocketTSLM: Deploying an On-Device Soft-Prompt Time-Series Language Model for Privacy-Preserving Smartphone Reasoning The paper presents PocketTSLM, deploying an on-device soft-prompt time-series language model on smartphones for personal health data. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Focuses on deployment efficiency and privacy architecture rather than safety guardrails or jailbreaks. | |
| Hybrid Generative-Tagging Decoding for Efficient Product Attribute-Value Extraction The paper introduces a hybrid generative-tagging architecture to improve efficiency and coverage in product attribute and value extraction. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Focuses on extraction throughput and accuracy rather than safety risks or guardrails. | |
| Ranking and Partitioning of Attention Heads for Sparse and Continual Reinforcement Learning Fine-tuning The paper studies ranking and partitioning attention heads by gradient norm for sparse and continual reinforcement learning fine-tuning. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Addresses catastrophic forgetting in training rather than safety guardrails or jailbreaks. | |
| OpCredit: Operation-Aware Policy Learning for Agentic Knowledge Graph Question Answering The paper proposes OpCredit, a policy learning framework that assigns individual operation credits for agentic knowledge graph question answering. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Focuses on question-answering accuracy rather than deployment safety, jailbreaks, or guardrails. | |
| Emotion Is Not External Knowledge: Internal Attention Signals for Affective State Reasoning The paper presents an internal attention boosting method to improve how language models infer implicit conversational affective states. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Studies emotion classification accuracy rather than AI deployment safety or jailbreak prevention. | |
| RenoLayout: Unified Floorplan Renovation and Furniture Layout for Constraint-Aware Residential Design The paper introduces RenoLayout to generate residential renovation floorplans and furniture layouts under structural and physical constraints. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Deals with architectural design synthesis rather than AI safety or guardrails. | |
| DPA-MoE: A Runtime Precision Control Framework for MoE Inference Introduces a dynamic precision control framework that allocates weight chunks to mixture-of-experts models during inference. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Focuses on serving throughput and quantization rather than AI safety or guardrails. | |
| Calibration Language Shapes Block-Influence Pruning in Multilingual Language Models Analyzes how calibration language choices affect block-influence depth pruning rankings in multilingual language models. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Studies structured pruning mechanics rather than deployment safety, guardrails, or jailbreaks. | |
| Beyond Invariance: Factor-wise Selective Treatment of Shifts for Multimodal Learning Proposes a factor-wise representation framework to selectively treat different distribution shifts in multimodal machine learning. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Focuses on multimodal classification robustness rather than safety, red-teaming, or guardrails. | |
| Identifying Effective Strategies in Prompt Evolution for Stress and Emotion Prediction Evaluates prompt optimization methods and identifies effective prompting strategies for social media stress and emotion classification. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Studies emotion classification accuracy rather than deployment guardrails or safety risks. | |
| HyCo: Hybrid Drafting for Cost-Efficient Mobile-Cloud Speculative Decoding Introduces a hybrid speculative decoding framework splitting draft tasks between mobile neural processing units and cloud servers. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Focuses on mobile inference efficiency rather than safety, alignment, or hallucination prevention. | |
| SITAR: A Situational Distortion Metric for Automatic Speech Recognition Evaluation Develops a lightweight situational distortion metric to evaluate speech recognition errors based on practical conversational severity. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Focuses on transcription scoring accuracy rather than deployment safety or jailbreak prevention. | |
| Translation Circuit Discovery and Targeted Optimization for Low-Resource Multilingual Translation Identifies translation circuits in multilingual models and optimizes fewer than five percent of parameters for low-resource translation. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Investigates mechanistic interpretability for translation without focusing on safety or jailbreaks. | |
| OPRD: On-Policy Reminder Distillation for Long-Horizon Coding Agents Uses multi-scope reminders and on-policy distillation to help coding agents maintain constraints over long software engineering interactions. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Focuses on coding task performance rather than safety, guardrails, or alignment risks. | |
| Writability and the Pathway: A Controlled Study of Knowledge Injection into Addressed Memory Conducts a controlled evaluation of direct knowledge injection into addressed conditional memory modules across small and production models. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Investigates internal memory writing mechanisms rather than deployment safety, jailbreaks, or guardrails. | |
| Waypoint-OPD: Transferring RL-Induced Policy Shifts via Gated On-Policy Distillation Gates reinforcement learning policy-shift distillation using intermediate checkpoints to ensure student models only learn consistent teacher shifts. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Addresses policy distillation performance rather than safety guardrails or attacks. | |
| Enhancing Multimodal RLVR with Pairwise Token Contrast and Adaptive Trajectory Confidence Improves reinforcement learning credit assignment in multimodal reasoning using pairwise token contrast and adaptive trajectory confidence. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Focuses on reasoning accuracy rather than deployment safety or jailbreak prevention. | |
| Diffusion Language Models as Open Quantum Systems Frames diffusion language models as open quantum systems to enable text generation conditioned on superpositions of prompts and thoughts. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Does not investigate AI safety guardrails, attacks, or robustness. | |
| No Single Tokenizer Feature Reliably Predicts Downstream Language Model Performance Shows across 200 models that proposed intrinsic tokenizer metrics do not reliably predict downstream language model performance. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Analyzes tokenizer predictability, not safety vulnerabilities or jailbreaks. | |
| Data Strategies for Large Language Models: A Unified Perspective on the Training Lifecycle Reviews and formalizes data selection, weighting, and scheduling policies across the pretraining and post-training lifecycle of language models. Preprint | AI safety and reliability in deployment via arxiv · 17 Oct, 06:17 | Not relevant Surveys general training data strategies rather than alignment or guardrails. |
961–1,020 of 3,579← PreviousNext →