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
ChemExplain: A Multimodal Benchmark for Explanatory Chemical Reasoning Presents a multimodal benchmark testing whether models can explain observed chemical effects by linking structural features to reaction mechanisms. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:17	Not relevant Focuses on chemistry molecular diagrams, not industrial inspection.
ChemExplain: A Multimodal Benchmark for Explanatory Chemical Reasoning Presents a multimodal benchmark testing whether models can explain observed chemical effects by linking structural features to reaction mechanisms. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Benchmark paper for chemistry reasoning, not autonomous LLM agents.
ChemExplain: A Multimodal Benchmark for Explanatory Chemical Reasoning Presents a multimodal benchmark testing whether models can explain observed chemical effects by linking structural features to reaction mechanisms. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Chemistry QA benchmark, not AI safety or jailbreak prevention.
FreeSolo: Linear-time Sparse Attention with Zero-Shot Length Extrapolation Introduces a native sparse attention mechanism that combines sliding windows and keypoint selection for linear-time context extrapolation. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Focuses on architectural design, not benchmark creation or evaluation.
FreeSolo: Linear-time Sparse Attention with Zero-Shot Length Extrapolation Introduces a native sparse attention mechanism that combines sliding windows and keypoint selection for linear-time context extrapolation. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Proposes a sparse attention mechanism, not agent architectures.
FreeSolo: Linear-time Sparse Attention with Zero-Shot Length Extrapolation Introduces a native sparse attention mechanism that combines sliding windows and keypoint selection for linear-time context extrapolation. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Focuses on attention length extrapolation, not deployment safety.
Size Generalization for Diffusion Routing Solvers Uses reinforcement learning on large routing instances to help diffusion-based solvers generalize beyond their training instance sizes. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Focuses on TSP/CVRP routing solvers, not AI safety.
Size Generalization for Diffusion Routing Solvers Uses reinforcement learning on large routing instances to help diffusion-based solvers generalize beyond their training instance sizes. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Addresses diffusion combinatorial optimization solvers, not LLM agents.
Size Generalization for Diffusion Routing Solvers Uses reinforcement learning on large routing instances to help diffusion-based solvers generalize beyond their training instance sizes. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Covers combinatorial optimization algorithms, not LLM evaluation.
Deleting Code Without Losing Knowledge: Regenerative Software for Python Libraries Applies regenerative software primitives so programs can be deleted and rebuilt by agents without losing architectural intent and knowledge. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Focuses on software architecture patterns, not AI safety guardrails.
Deleting Code Without Losing Knowledge: Regenerative Software for Python Libraries Applies regenerative software primitives so programs can be deleted and rebuilt by agents without losing architectural intent and knowledge. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Describes a software engineering methodology, not benchmark evaluation.
Releasing the Inner Jev: A Simple Recipe for Open Decision Models The paper demonstrates recipes to extract single-pass probability decisions directly from open instruction-tuned language models. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Studies single-forward-pass decision heads, not multi-step agent planning.
Releasing the Inner Jev: A Simple Recipe for Open Decision Models The paper demonstrates recipes to extract single-pass probability decisions directly from open instruction-tuned language models. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Covers logit classification and capability metrics, not deployment guardrails.
AutoMARS: Automatic and Scalable Multi-Agent Large Language Model Framework for Supply Chain Risk Assessment The paper proposes a multi-agent framework to coordinate analytical tools and evaluate multi-factor supply chain risks across diverse locations. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Domain risk assessment tool, not AI safety and reliability.
AutoMARS: Automatic and Scalable Multi-Agent Large Language Model Framework for Supply Chain Risk Assessment The paper proposes a multi-agent framework to coordinate analytical tools and evaluate multi-factor supply chain risks across diverse locations. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Focuses on multi-agent application design, not benchmark development.
Compute-optimal data scaling for neural surrogates via multi-fidelity training The paper demonstrates multi-fidelity training to optimize data generation compute and error scaling for neural PDE surrogates. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Covers scientific simulator scaling, not LLM evaluation benchmarks.
Compute-optimal data scaling for neural surrogates via multi-fidelity training The paper demonstrates multi-fidelity training to optimize data generation compute and error scaling for neural PDE surrogates. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Focuses on numerical PDE simulation accuracy, not AI safety.
Compute-optimal data scaling for neural surrogates via multi-fidelity training The paper demonstrates multi-fidelity training to optimize data generation compute and error scaling for neural PDE surrogates. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Studies surrogate models for PDEs, not LLM agents.
IsoVRBench: Video Reward Modeling Without Human Feedback The paper presents a benchmark of synthetic paired comparisons to evaluate video reward models across specific visual and motion dimensions. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Addresses video generation preference evaluation, not deployment safety.
IsoVRBench: Video Reward Modeling Without Human Feedback The paper presents a benchmark of synthetic paired comparisons to evaluate video reward models across specific visual and motion dimensions. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:17	Not relevant Studies generative video reward modeling, not industrial defect detection.
IsoVRBench: Video Reward Modeling Without Human Feedback The paper presents a benchmark of synthetic paired comparisons to evaluate video reward models across specific visual and motion dimensions. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Focuses on video reward models, not autonomous planning agents.
From Conceptual EneMonIE to an Executable Backend: A Modular Architecture for Energy Information Extractionand Grounded Question Answering The paper describes a modular software backend architecture for extracting energy domain information and answering grounded questions from web sources. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Software system implementation, not evaluation benchmarks or methodology.
From Conceptual EneMonIE to an Executable Backend: A Modular Architecture for Energy Information Extractionand Grounded Question Answering The paper describes a modular software backend architecture for extracting energy domain information and answering grounded questions from web sources. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Focuses on pipeline engineering, not model compression or efficiency.
From Conceptual EneMonIE to an Executable Backend: A Modular Architecture for Energy Information Extractionand Grounded Question Answering The paper describes a modular software backend architecture for extracting energy domain information and answering grounded questions from web sources. Preprint	Multimodal models via arxiv · 15 Oct, 06:17	Not relevant Deals with textual data and web metadata, not multimodality.
From Conceptual EneMonIE to an Executable Backend: A Modular Architecture for Energy Information Extractionand Grounded Question Answering The paper describes a modular software backend architecture for extracting energy domain information and answering grounded questions from web sources. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Covers information extraction pipeline rather than AI safety guardrails.
From Conceptual EneMonIE to an Executable Backend: A Modular Architecture for Energy Information Extractionand Grounded Question Answering The paper describes a modular software backend architecture for extracting energy domain information and answering grounded questions from web sources. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Focuses on backend data pipeline architecture, not autonomous agentic workflows.
From Conceptual EneMonIE to an Executable Backend: A Modular Architecture for Energy Information Extractionand Grounded Question Answering The paper describes a modular software backend architecture for extracting energy domain information and answering grounded questions from web sources. Preprint	LLMs for code via arxiv · 15 Oct, 06:17	Not relevant Application backend architecture, not LLMs for software development.
Does a Governed Confidence Score Know When a Text-to-SQL Answer Is Right? An Exploratory Study on BIRD and a Pre-registered Confirmation on Spider The paper audits an enterprise confidence scoring mechanism for text-to-SQL models on the BIRD and Spider benchmarks. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Studies confidence scoring in text-to-SQL systems, not multi-step agents.
RL Post-Training Diffusion Language Models Along Their Own Generation Paths The paper presents a reinforcement learning post-training method that scores intermediate generation trajectories in diffusion language models. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Focuses on RL post-training for diffusion language models, not agents.
RL Post-Training Diffusion Language Models Along Their Own Generation Paths The paper presents a reinforcement learning post-training method that scores intermediate generation trajectories in diffusion language models. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Proposes an RL training algorithm, not an evaluation benchmark.
RL Post-Training Diffusion Language Models Along Their Own Generation Paths The paper presents a reinforcement learning post-training method that scores intermediate generation trajectories in diffusion language models. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Studies training consistency for diffusion LMs, not deployment safety.
Measuring Learned Equivariance: An Information Theoretic Approach The paper proposes information-theoretic tests to measure learned equivariance in intermediate neural network representations without knowing explicit group transformations. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Analyzes representation symmetries, not autonomous planning agents.
Measuring Learned Equivariance: An Information Theoretic Approach The paper proposes information-theoretic tests to measure learned equivariance in intermediate neural network representations without knowing explicit group transformations. Preprint	Small and efficient models via arxiv · 15 Oct, 06:17	Not relevant Analyzes representation equivariance rather than compression or serving.
Measuring Learned Equivariance: An Information Theoretic Approach The paper proposes information-theoretic tests to measure learned equivariance in intermediate neural network representations without knowing explicit group transformations. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Focuses on geometric representation theory, not deployment safety risks.
Measuring Learned Equivariance: An Information Theoretic Approach The paper proposes information-theoretic tests to measure learned equivariance in intermediate neural network representations without knowing explicit group transformations. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Deals with representation invariance metrics rather than LLM benchmarks.
Reasoning for Propaganda Detection via Label-Guided CoT Optimisation The paper presents a reasoning-conditioned framework that refines chain-of-thought explanations to improve propaganda detection without explanation annotations. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Proposes a classification framework rather than benchmark evaluations.
Reasoning for Propaganda Detection via Label-Guided CoT Optimisation The paper presents a reasoning-conditioned framework that refines chain-of-thought explanations to improve propaganda detection without explanation annotations. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Focuses on text classification rather than system deployment safety.
Reasoning for Propaganda Detection via Label-Guided CoT Optimisation The paper presents a reasoning-conditioned framework that refines chain-of-thought explanations to improve propaganda detection without explanation annotations. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Presents a propaganda classification pipeline, not an autonomous agent.
Phase-Conditioned Affordance Fields for SE(3)-Equivariant Visuomotor Policies The paper introduces phase-conditioned affordance fields to improve equivariant visuomotor policy learning for robotic manipulation. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:17	Not relevant Covers general robotic manipulation rather than industrial defect inspection.
Phase-Conditioned Affordance Fields for SE(3)-Equivariant Visuomotor Policies The paper introduces phase-conditioned affordance fields to improve equivariant visuomotor policy learning for robotic manipulation. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Deals with robotic affordance fields rather than deployment safety.
Phase-Conditioned Affordance Fields for SE(3)-Equivariant Visuomotor Policies The paper introduces phase-conditioned affordance fields to improve equivariant visuomotor policy learning for robotic manipulation. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Introduces a robotic policy method rather than LLM benchmarks.
Phase-Conditioned Affordance Fields for SE(3)-Equivariant Visuomotor Policies The paper introduces phase-conditioned affordance fields to improve equivariant visuomotor policy learning for robotic manipulation. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Focuses on robot manipulation policies rather than LLM-based autonomous agents.
PhysCodeGym: Benchmarking Coding Agents for Physically Verifiable 4D Reconstruction The paper introduces a benchmark to evaluate coding agents building executable physical simulations from video observations using Blender. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Focuses on simulation benchmarks rather than deployment safety risks.
Towards Resilience: Slice-Driven Orchestration of Mobile IoMT in 6G-Edge-Continuum The paper introduces an adaptive network-slicing orchestration framework for medical IoT applications operating across edge-computing infrastructure. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Addresses telecommunications QoS rather than AI safety guardrails.
Towards Resilience: Slice-Driven Orchestration of Mobile IoMT in 6G-Edge-Continuum The paper introduces an adaptive network-slicing orchestration framework for medical IoT applications operating across edge-computing infrastructure. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Deals with network slicing in edge computing, not AI agents.
Towards Resilience: Slice-Driven Orchestration of Mobile IoMT in 6G-Edge-Continuum The paper introduces an adaptive network-slicing orchestration framework for medical IoT applications operating across edge-computing infrastructure. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Focuses on network orchestration rather than LLM evaluation.
The Agent Knew: Internal States Forecast Unauthorized Actions and Stop Them, and the Lies That Follow The paper demonstrates that internal activation states can forecast and block unauthorized actions and deceptive reports by language model agents. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Focuses on action forecasting and gating rather than benchmark methodology.
Reliable Text-to-SQL in Resource-Constrained Settings with Small LLMs via Iterative Retry Workflows The paper evaluates small language models for text-to-SQL tasks and tests lightweight iterative retry workflows to improve query accuracy. Peer-reviewed	AI agents via arxiv · 15 Oct, 06:17	Not relevant Focuses on text-to-SQL prompting rather than autonomous tool agents.
Reliable Text-to-SQL in Resource-Constrained Settings with Small LLMs via Iterative Retry Workflows The paper evaluates small language models for text-to-SQL tasks and tests lightweight iterative retry workflows to improve query accuracy. Peer-reviewed	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Focuses on SQL query accuracy rather than AI deployment safety.
Reliable Text-to-SQL in Resource-Constrained Settings with Small LLMs via Iterative Retry Workflows The paper evaluates small language models for text-to-SQL tasks and tests lightweight iterative retry workflows to improve query accuracy. Peer-reviewed	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Evaluates existing benchmarks rather than creating new testing frameworks.
StartLux-Decision: AutoThink Decision Models with Reinforcement Learning for Calibrated Deliberative Decisions The paper introduces decision models that balance fast single-pass predictions with adaptive reasoning for uncertain cases using reinforcement learning. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Proposes a model training framework, not an evaluation benchmark.
StartLux-Decision: AutoThink Decision Models with Reinforcement Learning for Calibrated Deliberative Decisions The paper introduces decision models that balance fast single-pass predictions with adaptive reasoning for uncertain cases using reinforcement learning. Preprint	Retrieval-augmented generation (RAG) via arxiv · 15 Oct, 06:17	Not relevant Presents adaptive reasoning models without retrieval-augmented components.
StartLux-Decision: AutoThink Decision Models with Reinforcement Learning for Calibrated Deliberative Decisions The paper introduces decision models that balance fast single-pass predictions with adaptive reasoning for uncertain cases using reinforcement learning. Preprint	Multimodal models via arxiv · 15 Oct, 06:17	Not relevant Focuses on text decision reasoning, not multimodal inputs.
StartLux-Decision: AutoThink Decision Models with Reinforcement Learning for Calibrated Deliberative Decisions The paper introduces decision models that balance fast single-pass predictions with adaptive reasoning for uncertain cases using reinforcement learning. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Focuses on deliberative model reasoning rather than autonomous tool-using agents.
StartLux-Decision: AutoThink Decision Models with Reinforcement Learning for Calibrated Deliberative Decisions The paper introduces decision models that balance fast single-pass predictions with adaptive reasoning for uncertain cases using reinforcement learning. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Focuses on decision calibration rather than deployment safety guardrails.
StartLux-Decision: AutoThink Decision Models with Reinforcement Learning for Calibrated Deliberative Decisions The paper introduces decision models that balance fast single-pass predictions with adaptive reasoning for uncertain cases using reinforcement learning. Preprint	LLMs for code via arxiv · 15 Oct, 06:17	Not relevant Covers general decision reasoning rather than software code development.
Beyond Successful Play: Diagnosing Opponent Modeling in Large Language Models The paper introduces a diagnostic framework to test whether language model agents genuinely infer opponent beliefs rather than following superficial behavioral patterns. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Focuses on opponent modeling rather than deployment safety guardrails.
Joint Embedding with Dynamic Integration network (JEDI-net) The paper proposes clustering observations by feature importance profiles to incorporate latent factor representations into deep learning models. Preprint	AI agents via arxiv · 15 Oct, 06:17	Not relevant Presents a deep learning architecture, not multi-step agents.
Joint Embedding with Dynamic Integration network (JEDI-net) The paper proposes clustering observations by feature importance profiles to incorporate latent factor representations into deep learning models. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:17	Not relevant Deals with feature clustering rather than deployment safety guardrails.
Joint Embedding with Dynamic Integration network (JEDI-net) The paper proposes clustering observations by feature importance profiles to incorporate latent factor representations into deep learning models. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:17	Not relevant Proposes model architecture rather than LLM evaluation benchmarks.

13,261–13,320 of 17,313← Previous Next →