Every paper checked — nothing hidden

All papers

Every paper checked for brief #4 of 12–19 Oct 2026 — nothing is hidden. Brief papers first.

All17,313 In brief24 More (on-topic)3,333 Not relevant13,754 Not assessed87 Older352 Could not confirm93 Blogs with no paper266 All blogs284

Paper	Topic	Relevance
RUBATO: Decoupled Clocks Allow Full-Duplex Models to Think as Fast as They Can The authors propose decoupled positional encodings enabling full-duplex speech models to run internal thinking streams at hardware-adaptive speeds. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Presents speech architecture encoding, not multi-step tool-using agents.
RUBATO: Decoupled Clocks Allow Full-Duplex Models to Think as Fast as They Can The authors propose decoupled positional encodings enabling full-duplex speech models to run internal thinking streams at hardware-adaptive speeds. Preprint	LLMs for code via arxiv · 15 Oct, 06:17	Not relevant Focuses on speech models rather than code generation assistants.
ConicFlow: constrained Inference-time optimal control for flow matching models The paper introduces a receding-horizon optimal control algorithm for constrained sampling in flow matching generative models. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not focus on small models, distillation, or quantization.
ConicFlow: constrained Inference-time optimal control for flow matching models The paper introduces a receding-horizon optimal control algorithm for constrained sampling in flow matching generative models. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Addresses generative trajectory optimization rather than AI deployment safety.
ConicFlow: constrained Inference-time optimal control for flow matching models The paper introduces a receding-horizon optimal control algorithm for constrained sampling in flow matching generative models. Preprint	Multimodal models via arxiv · 15 Oct, 06:17	Not relevant Concerns generative flow matching trajectories rather than multimodal reasoning.
ConicFlow: constrained Inference-time optimal control for flow matching models The paper introduces a receding-horizon optimal control algorithm for constrained sampling in flow matching generative models. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Presents sampling control methods, not LLM evaluation benchmarks.
ConicFlow: constrained Inference-time optimal control for flow matching models The paper introduces a receding-horizon optimal control algorithm for constrained sampling in flow matching generative models. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Addresses optimal control in flow matching, not agent workflows.
ConicFlow: constrained Inference-time optimal control for flow matching models The paper introduces a receding-horizon optimal control algorithm for constrained sampling in flow matching generative models. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:17	Not relevant Discusses mathematical flow sampling, not industrial vision defect detection.
MvFFN: Multi-view Floor plan Feed-forward Network for Unposed Wide-Baseline Panorama Layout Reconstruction The authors present a feed-forward network that aggregates local wall predictions to reconstruct floor plans from sparse unposed panoramas. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Focuses on 3D layout reconstruction, not autonomous agents.
MvFFN: Multi-view Floor plan Feed-forward Network for Unposed Wide-Baseline Panorama Layout Reconstruction The authors present a feed-forward network that aggregates local wall predictions to reconstruct floor plans from sparse unposed panoramas. Preprint	Multimodal models via arxiv · 15 Oct, 06:17	Not relevant Focuses on geometric computer vision rather than multimodal vision-language.
MvFFN: Multi-view Floor plan Feed-forward Network for Unposed Wide-Baseline Panorama Layout Reconstruction The authors present a feed-forward network that aggregates local wall predictions to reconstruct floor plans from sparse unposed panoramas. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Concerns floor-plan layout reconstruction, not AI deployment safety.
MvFFN: Multi-view Floor plan Feed-forward Network for Unposed Wide-Baseline Panorama Layout Reconstruction The authors present a feed-forward network that aggregates local wall predictions to reconstruct floor plans from sparse unposed panoramas. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Evaluates 3D layout prediction rather than LLM benchmarks.
MvFFN: Multi-view Floor plan Feed-forward Network for Unposed Wide-Baseline Panorama Layout Reconstruction The authors present a feed-forward network that aggregates local wall predictions to reconstruct floor plans from sparse unposed panoramas. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:17	Not relevant Reconstructs floor plans from panoramas, not industrial defect inspection.
MvFFN: Multi-view Floor plan Feed-forward Network for Unposed Wide-Baseline Panorama Layout Reconstruction The authors present a feed-forward network that aggregates local wall predictions to reconstruct floor plans from sparse unposed panoramas. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not study model compression, distillation, or efficient inference.
EV-JEPA: Elastic V-JEPA for Any Device and Cross-Device Inference The paper presents an elastic vision architecture that dynamically reconfigures width and supports partitionable inference across heterogeneous edge devices. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Does not propose benchmarks or evaluations for LLMs.
EV-JEPA: Elastic V-JEPA for Any Device and Cross-Device Inference The paper presents an elastic vision architecture that dynamically reconfigures width and supports partitionable inference across heterogeneous edge devices. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Focuses on execution efficiency rather than safety or guardrails.
EV-JEPA: Elastic V-JEPA for Any Device and Cross-Device Inference The paper presents an elastic vision architecture that dynamically reconfigures width and supports partitionable inference across heterogeneous edge devices. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:17	Not relevant Presents generic vision architecture rather than industrial defect inspection.
EV-JEPA: Elastic V-JEPA for Any Device and Cross-Device Inference The paper presents an elastic vision architecture that dynamically reconfigures width and supports partitionable inference across heterogeneous edge devices. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Proposes vision model architectures, not autonomous multi-step agents.
TraceSampler: Making Sampling Errors Predictable and Correctable for Accelerated Generative Sampling The paper introduces a post-training residual correction method to predict and correct endpoint errors during low-step generative model sampling. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Deals with continuous-time generative sampling, not autonomous agents.
TraceSampler: Making Sampling Errors Predictable and Correctable for Accelerated Generative Sampling The paper introduces a post-training residual correction method to predict and correct endpoint errors during low-step generative model sampling. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:17	Not relevant Focuses on general image generation rather than industrial inspection.
TraceSampler: Making Sampling Errors Predictable and Correctable for Accelerated Generative Sampling The paper introduces a post-training residual correction method to predict and correct endpoint errors during low-step generative model sampling. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Studies generative sampling acceleration, not safety, guardrails, or alignment.
TraceSampler: Making Sampling Errors Predictable and Correctable for Accelerated Generative Sampling The paper introduces a post-training residual correction method to predict and correct endpoint errors during low-step generative model sampling. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Proposes sampling error correction, not LLM benchmark evaluation.
TraceSampler: Making Sampling Errors Predictable and Correctable for Accelerated Generative Sampling The paper introduces a post-training residual correction method to predict and correct endpoint errors during low-step generative model sampling. Preprint	Multimodal models via arxiv · 15 Oct, 06:17	Not relevant Concerns image diffusion sampling rather than multimodal language reasoning.
Which Metadata Matters for LLM Tool Selection in MCP? Evidence from Controlled Factor-Level Interventions The authors run controlled experiments showing how parameter schema differences in metadata alter language model tool selection. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Not related to small models, quantization, or compute efficiency.
Which Metadata Matters for LLM Tool Selection in MCP? Evidence from Controlled Factor-Level Interventions The authors run controlled experiments showing how parameter schema differences in metadata alter language model tool selection. Preprint	LLMs for code via arxiv · 15 Oct, 06:17	Not relevant Addresses tool selection in MCP rather than code generation or repair.
Which Metadata Matters for LLM Tool Selection in MCP? Evidence from Controlled Factor-Level Interventions The authors run controlled experiments showing how parameter schema differences in metadata alter language model tool selection. Preprint	Multimodal models via arxiv · 15 Oct, 06:17	Not relevant Examines text-based tool metadata, not multimodal vision-language models.
Which Metadata Matters for LLM Tool Selection in MCP? Evidence from Controlled Factor-Level Interventions The authors run controlled experiments showing how parameter schema differences in metadata alter language model tool selection. Preprint	Retrieval-augmented generation (RAG) via arxiv · 15 Oct, 06:17	Not relevant Focuses on tool selection metadata, not retrieval-augmented generation pipelines.
Which Metadata Matters for LLM Tool Selection in MCP? Evidence from Controlled Factor-Level Interventions The authors run controlled experiments showing how parameter schema differences in metadata alter language model tool selection. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Studies tool schema sensitivity rather than safety, red-teaming, or guardrails.
Which Metadata Matters for LLM Tool Selection in MCP? Evidence from Controlled Factor-Level Interventions The authors run controlled experiments showing how parameter schema differences in metadata alter language model tool selection. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Evaluates tool metadata impact rather than general LLM benchmarking.
Natural Code Transformation Confuses Agents in Vulnerability Analysis The study reports that natural code transformations confuse vulnerability analysis agents, proposing functional changes as a potential defensive strategy. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Explores agent attack defense rather than general LLM evaluation benchmarking.
Natural Code Transformation Confuses Agents in Vulnerability Analysis The study reports that natural code transformations confuse vulnerability analysis agents, proposing functional changes as a potential defensive strategy. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not address model compression, quantization, or efficient inference.
Actionable Meta-Requirements for AI Robustness Auditing The paper establishes and validates a set of meta-requirements to guide the design of frameworks for auditing AI system robustness. Preprint	LLMs for code via arxiv · 15 Oct, 06:17	Not relevant Not focused on code generation or program repair tools.
Actionable Meta-Requirements for AI Robustness Auditing The paper establishes and validates a set of meta-requirements to guide the design of frameworks for auditing AI system robustness. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Auditing requirements for general AI, not interactive LLM agent workflows.
Actionable Meta-Requirements for AI Robustness Auditing The paper establishes and validates a set of meta-requirements to guide the design of frameworks for auditing AI system robustness. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Addresses governance auditing requirements rather than LLM benchmark evaluation methodology.
When Pleasing the User Compromises Care: AI Sycophancy in Caregiver-Mediated Healthcare The paper benchmarks how caregiver-directed language models compromise medical advice when users express clinically inferior care preferences. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Studies conversational advice drift rather than multi-step tool-using agents.
DexFLEX: Contact-Aware Foundation Controller for Command-Guided Dexterity The authors present a foundation controller that converts motion drafts into contact-consistent joint commands for dexterous robotic hands. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not target model quantization, distillation, or efficient inference serving.
DexFLEX: Contact-Aware Foundation Controller for Command-Guided Dexterity The authors present a foundation controller that converts motion drafts into contact-consistent joint commands for dexterous robotic hands. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Focuses on low-level robotic hand control, not language model agents.
DexFLEX: Contact-Aware Foundation Controller for Command-Guided Dexterity The authors present a foundation controller that converts motion drafts into contact-consistent joint commands for dexterous robotic hands. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Addresses physical grasp stability rather than software AI safety or alignment.
DexFLEX: Contact-Aware Foundation Controller for Command-Guided Dexterity The authors present a foundation controller that converts motion drafts into contact-consistent joint commands for dexterous robotic hands. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Introduces a robotic motion controller rather than an evaluation benchmark.
DexFLEX: Contact-Aware Foundation Controller for Command-Guided Dexterity The authors present a foundation controller that converts motion drafts into contact-consistent joint commands for dexterous robotic hands. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:17	Not relevant Addresses dexterous hand manipulation, not industrial defect detection or visual inspection.
ACED-Bench: Evaluating How LLM Agents Acquire and Act on Causal Evidence The paper introduces a neuro-symbolic benchmark to evaluate whether language model agents actively investigate environments and make causal decisions. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Focuses on causal experimentation benchmarks rather than safety, red-teaming, or guardrails.
ANASSA: From High-Level Framework to Low-Level Architecture for Spatial Agentic AI The paper refines a high-level spatial AI orchestration framework into an implementation-focused low-level architecture for governed agent execution. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Proposes a system architecture design rather than an evaluation benchmark.
ANASSA: From High-Level Framework to Low-Level Architecture for Spatial Agentic AI The paper refines a high-level spatial AI orchestration framework into an implementation-focused low-level architecture for governed agent execution. Preprint	LLMs for code via arxiv · 15 Oct, 06:17	Not relevant Presents an agent architecture for geospatial data, not code generation or repair.
Deep learning from a control vantage point The paper reviews deep learning architectures and training dynamics through the lens of continuous-time nonlinear control theory. Peer-reviewed	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Theoretical overview of architectures, not benchmark evaluation or testing methodology.
Deep learning from a control vantage point The paper reviews deep learning architectures and training dynamics through the lens of continuous-time nonlinear control theory. Peer-reviewed	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not address model compression, pruning, quantization, or serving optimizations.
Deep learning from a control vantage point The paper reviews deep learning architectures and training dynamics through the lens of continuous-time nonlinear control theory. Peer-reviewed	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Does not study safety, hallucination, prompt injection, or deployment guardrails.
Deep learning from a control vantage point The paper reviews deep learning architectures and training dynamics through the lens of continuous-time nonlinear control theory. Peer-reviewed	AI agents via arxiv · 15 Oct, 06:17	Not relevant Surveys control theory connections to deep learning architectures, not agent workflows.
Measuring the Inference Energy Efficiency of Open-Weight LLMs across Benchmarks, Reasoning Modes, and Hardware The paper measures inference energy efficiency across 86 open-weight language models, showing large efficiency gaps among models of similar capability. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Focuses on hardware energy consumption rather than safety, alignment, or jailbreaks.
Measuring the Inference Energy Efficiency of Open-Weight LLMs across Benchmarks, Reasoning Modes, and Hardware The paper measures inference energy efficiency across 86 open-weight language models, showing large efficiency gaps among models of similar capability. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Studies energy efficiency of LLMs, not autonomous agent architectures.
RealityBench: Simulated Environments to Evaluate a Vision Language Model's Ability to Document Human Physical Work The paper presents a simulated benchmark across eleven unconstrained environments to evaluate vision-language models on documenting real-world human physical work. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Focuses on physical work documentation accuracy rather than deployment safety guardrails.
RealityBench: Simulated Environments to Evaluate a Vision Language Model's Ability to Document Human Physical Work The paper presents a simulated benchmark across eleven unconstrained environments to evaluate vision-language models on documenting real-world human physical work. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Evaluates vision-language model understanding, not multi-step agent planning or tool use.
RealityBench: Simulated Environments to Evaluate a Vision Language Model's Ability to Document Human Physical Work The paper presents a simulated benchmark across eleven unconstrained environments to evaluate vision-language models on documenting real-world human physical work. Preprint	LLMs for code via arxiv · 15 Oct, 06:17	Not relevant Does not involve software code generation or developer assistants.
RealityBench: Simulated Environments to Evaluate a Vision Language Model's Ability to Document Human Physical Work The paper presents a simulated benchmark across eleven unconstrained environments to evaluate vision-language models on documenting real-world human physical work. Preprint	Retrieval-augmented generation (RAG) via arxiv · 15 Oct, 06:17	Not relevant Does not involve retrieval-augmented generation or external document grounding.
RealityBench: Simulated Environments to Evaluate a Vision Language Model's Ability to Document Human Physical Work The paper presents a simulated benchmark across eleven unconstrained environments to evaluate vision-language models on documenting real-world human physical work. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Does not investigate model compression, quantization, or inference acceleration.
SAT-RL: Training Language Models for Logical Reasoning with Symbolic Solvers The paper presents a reinforcement learning framework that trains language models to solve complex logical reasoning tasks using interactive symbolic solvers. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Presents a training method with solvers rather than an evaluation framework.
SAT-RL: Training Language Models for Logical Reasoning with Symbolic Solvers The paper presents a reinforcement learning framework that trains language models to solve complex logical reasoning tasks using interactive symbolic solvers. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Focuses on logical reasoning accuracy rather than deployment safety or jailbreaks.
Memory Without Curation: Learning Multi-Stream Latent Memory for Robot Policies The paper proposes an uncurated multi-stream latent memory framework that augments vision-language-action robotic policies while maintaining low inference latency. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Focuses on low-level robotic manipulation control rather than LLM-based agents.
Memory Without Curation: Learning Multi-Stream Latent Memory for Robot Policies The paper proposes an uncurated multi-stream latent memory framework that augments vision-language-action robotic policies while maintaining low inference latency. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Does not focus on safety guardrails, jailbreaks, or alignment monitoring.
Memory Without Curation: Learning Multi-Stream Latent Memory for Robot Policies The paper proposes an uncurated multi-stream latent memory framework that augments vision-language-action robotic policies while maintaining low inference latency. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Develops a latent memory framework for robots rather than general model compression.
Memory Without Curation: Learning Multi-Stream Latent Memory for Robot Policies The paper proposes an uncurated multi-stream latent memory framework that augments vision-language-action robotic policies while maintaining low inference latency. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Focuses on a policy architecture rather than general LLM evaluation benchmarks.

8,281–8,340 of 13,754← Previous Next →