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
How Inner Optimizers Shape Replica Scaling Studies how the choice of local optimizers affects model performance degradation when scaling distributed training replicas under fixed budgets. Preprint	AI agents via arxiv · 15 Oct, 15:06	Not relevant Concerns distributed pretraining optimizers rather than multi-step language model agents.
Revisiting Generative vs. Dense Retrieval: A Static Value Prior Reverses the Gap Shows that incorporating static item value into dense retrieval outperforms generative retrieval in e-commerce recommendation settings. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 15:06	Not relevant Focuses on retrieval algorithms rather than language model evaluation frameworks or leaderboards.
Revisiting Generative vs. Dense Retrieval: A Static Value Prior Reverses the Gap Shows that incorporating static item value into dense retrieval outperforms generative retrieval in e-commerce recommendation settings. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 15:06	Not relevant Deals with recommendation retrieval accuracy rather than AI safety or reliability.
Revisiting Generative vs. Dense Retrieval: A Static Value Prior Reverses the Gap Shows that incorporating static item value into dense retrieval outperforms generative retrieval in e-commerce recommendation settings. Preprint	AI agents via arxiv · 15 Oct, 15:06	Not relevant Focuses on candidate retrieval for recommendation rather than multi-step agent planning or tools.
Revisiting Generative vs. Dense Retrieval: A Static Value Prior Reverses the Gap Shows that incorporating static item value into dense retrieval outperforms generative retrieval in e-commerce recommendation settings. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on retrieval quality comparison rather than model compression or quantization techniques.
Plan or Retrieve First? A Scale Study of Tool Discovery Architectures Compares retrieve-then-plan and plan-then-retrieve strategies for tool discovery across varying tool registry complexities. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 15:06	Not relevant Focuses on tool retrieval efficiency rather than safety guardrails or jailbreak mitigation.
Plan or Retrieve First? A Scale Study of Tool Discovery Architectures Compares retrieve-then-plan and plan-then-retrieve strategies for tool discovery across varying tool registry complexities. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 15:06	Not relevant Focuses on agent tool discovery architectures rather than general benchmark validity or evaluation.
Beyond Aggregate Accuracy: Item-Level Transitions in Prompt Optimization Introduces a prompt optimization framework that monitors item-level regressions and recoveries rather than relying solely on aggregate accuracy. Preprint	Multimodal models via arxiv · 15 Oct, 15:06	Not relevant Investigates text-based prompt optimization without multimodal components.
Beyond Aggregate Accuracy: Item-Level Transitions in Prompt Optimization Introduces a prompt optimization framework that monitors item-level regressions and recoveries rather than relying solely on aggregate accuracy. Preprint	AI agents via arxiv · 15 Oct, 15:06	Not relevant Studies automatic prompt optimization dynamics rather than multi-step autonomous tool-using agents.
Beyond Aggregate Accuracy: Item-Level Transitions in Prompt Optimization Introduces a prompt optimization framework that monitors item-level regressions and recoveries rather than relying solely on aggregate accuracy. Preprint	LLMs for code via arxiv · 15 Oct, 15:06	Not relevant Studies prompt optimization across domains rather than code generation or software engineering tools.
Beyond Aggregate Accuracy: Item-Level Transitions in Prompt Optimization Introduces a prompt optimization framework that monitors item-level regressions and recoveries rather than relying solely on aggregate accuracy. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Examines prompt engineering metrics rather than model quantization, distillation, or serving.
Beyond Aggregate Accuracy: Item-Level Transitions in Prompt Optimization Introduces a prompt optimization framework that monitors item-level regressions and recoveries rather than relying solely on aggregate accuracy. Preprint	Retrieval-augmented generation (RAG) via arxiv · 15 Oct, 15:06	Not relevant Analyzes prompt optimization across benchmarks rather than external knowledge retrieval or RAG.
DexMimic: A Generalist Dexterous Hand-Object Tracking Policy via Value-Guided Reset Allocation Introduces a generalist policy that tracks diverse dexterous hand-object interactions using value-guided reset allocation during training. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 15:06	Not relevant Focuses on reinforcement learning for robotics rather than testing language models.
DexMimic: A Generalist Dexterous Hand-Object Tracking Policy via Value-Guided Reset Allocation Introduces a generalist policy that tracks diverse dexterous hand-object interactions using value-guided reset allocation during training. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 15:06	Not relevant Focuses on dexterous hand manipulation rather than industrial visual inspection or defect detection.
DexMimic: A Generalist Dexterous Hand-Object Tracking Policy via Value-Guided Reset Allocation Introduces a generalist policy that tracks diverse dexterous hand-object interactions using value-guided reset allocation during training. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 15:06	Not relevant Deals with motor control tracking rather than AI safety, hallucination, or alignment.
DexMimic: A Generalist Dexterous Hand-Object Tracking Policy via Value-Guided Reset Allocation Introduces a generalist policy that tracks diverse dexterous hand-object interactions using value-guided reset allocation during training. Preprint	AI agents via arxiv · 15 Oct, 15:06	Not relevant Covers physical robotic hand control rather than LLM-based planning or tool-use agents.
GroundGUI: GUI Grounding by High-Quality Data and Distillation-to-Reinforcement Learning Presents a large graphical user interface grounding dataset and a combined distillation and reinforcement learning training method for agents. Preprint	LLMs for code via arxiv · 15 Oct, 15:06	Not relevant Focuses on user interface grounding rather than source code generation or software engineering.
GroundGUI: GUI Grounding by High-Quality Data and Distillation-to-Reinforcement Learning Presents a large graphical user interface grounding dataset and a combined distillation and reinforcement learning training method for agents. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 15:06	Not relevant Addresses GUI referring expressions rather than safety guardrails, jailbreaking, or red teaming.
GroundGUI: GUI Grounding by High-Quality Data and Distillation-to-Reinforcement Learning Presents a large graphical user interface grounding dataset and a combined distillation and reinforcement learning training method for agents. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on grounding accuracy rather than model compression or efficient serving architectures.
GroundGUI: GUI Grounding by High-Quality Data and Distillation-to-Reinforcement Learning Presents a large graphical user interface grounding dataset and a combined distillation and reinforcement learning training method for agents. Preprint	Retrieval-augmented generation (RAG) via arxiv · 15 Oct, 15:06	Not relevant Deals with on-screen component grounding rather than external knowledge retrieval.
GroundGUI: GUI Grounding by High-Quality Data and Distillation-to-Reinforcement Learning Presents a large graphical user interface grounding dataset and a combined distillation and reinforcement learning training method for agents. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 15:06	Not relevant Introduces a dataset and model for GUI grounding rather than evaluating LLM benchmark reliability.
GroundGUI: GUI Grounding by High-Quality Data and Distillation-to-Reinforcement Learning Presents a large graphical user interface grounding dataset and a combined distillation and reinforcement learning training method for agents. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 15:06	Not relevant Pertains to computer screen GUIs rather than manufacturing or industrial defect inspection.
Towards a Data-Centric Understanding of Spatial Reasoning in Multimodal Large Language Models Surveys dataset design, construction methods, and evaluation protocols for improving spatial reasoning in multimodal language models. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 15:06	Not relevant Covers generic spatial reasoning rather than industrial inspection or manufacturing defect detection.
Towards a Data-Centric Understanding of Spatial Reasoning in Multimodal Large Language Models Surveys dataset design, construction methods, and evaluation protocols for improving spatial reasoning in multimodal language models. Preprint	Retrieval-augmented generation (RAG) via arxiv · 15 Oct, 15:06	Not relevant Does not address external document retrieval or retrieval-augmented generation pipelines.
Towards a Data-Centric Understanding of Spatial Reasoning in Multimodal Large Language Models Surveys dataset design, construction methods, and evaluation protocols for improving spatial reasoning in multimodal language models. Preprint	LLMs for code via arxiv · 15 Oct, 15:06	Not relevant Surveys spatial vision datasets without relevance to software code or program repair.
Towards a Data-Centric Understanding of Spatial Reasoning in Multimodal Large Language Models Surveys dataset design, construction methods, and evaluation protocols for improving spatial reasoning in multimodal language models. Preprint	AI agents via arxiv · 15 Oct, 15:06	Not relevant Surveys spatial dataset construction for vision models rather than autonomous agent systems.
Towards a Data-Centric Understanding of Spatial Reasoning in Multimodal Large Language Models Surveys dataset design, construction methods, and evaluation protocols for improving spatial reasoning in multimodal language models. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Surveys datasets and evaluation rather than model pruning, quantization, or compression.
Towards a Data-Centric Understanding of Spatial Reasoning in Multimodal Large Language Models Surveys dataset design, construction methods, and evaluation protocols for improving spatial reasoning in multimodal language models. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 15:06	Not relevant Examines spatial perception datasets rather than AI safety, red-teaming, or guardrails.
From Copying to Composition: How Language Models Develop In-Context Learning Analyzes how language models develop in-context learning by separating simple demonstration copying from multi-token rule composition across training checkpoints. Preprint	LLMs for code via arxiv · 15 Oct, 15:06	Not relevant Examines general prompt comprehension rather than software engineering or code generation.
From Copying to Composition: How Language Models Develop In-Context Learning Analyzes how language models develop in-context learning by separating simple demonstration copying from multi-token rule composition across training checkpoints. Preprint	Multimodal models via arxiv · 15 Oct, 15:06	Not relevant Investigates unimodal text language models without multimodal components.
From Copying to Composition: How Language Models Develop In-Context Learning Analyzes how language models develop in-context learning by separating simple demonstration copying from multi-token rule composition across training checkpoints. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on context learning dynamics rather than model compression or inference efficiency.
From Copying to Composition: How Language Models Develop In-Context Learning Analyzes how language models develop in-context learning by separating simple demonstration copying from multi-token rule composition across training checkpoints. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 15:06	Not relevant Explores mechanistic in-context learning rather than deployment safety, red teaming, or guardrails.
From Copying to Composition: How Language Models Develop In-Context Learning Analyzes how language models develop in-context learning by separating simple demonstration copying from multi-token rule composition across training checkpoints. Preprint	Retrieval-augmented generation (RAG) via arxiv · 15 Oct, 15:06	Not relevant Studies prompt-level in-context learning rather than external retrieval systems or RAG.
From Copying to Composition: How Language Models Develop In-Context Learning Analyzes how language models develop in-context learning by separating simple demonstration copying from multi-token rule composition across training checkpoints. Preprint	AI agents via arxiv · 15 Oct, 15:06	Not relevant Investigates foundational in-context learning mechanics rather than agentic planning or tools.
Genesis Steering: Better Activation Steering Through Training Data Steers language model activations using vectors derived directly from the model's own training data or inferred proxy data. Preprint	AI agents via arxiv · 15 Oct, 15:06	Not relevant Investigates activation steering vectors rather than multi-step agent planning or tool use.
Genesis Steering: Better Activation Steering Through Training Data Steers language model activations using vectors derived directly from the model's own training data or inferred proxy data. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Focuses on representation steering rather than model compression, quantization, or efficient serving.
Genesis Steering: Better Activation Steering Through Training Data Steers language model activations using vectors derived directly from the model's own training data or inferred proxy data. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 15:06	Not relevant Develops an activation steering technique rather than an evaluation framework or benchmark.
Unlocking Structured Pruning for Reasoning Language Models via Attention-Preserving MLP Trimming The paper proposes a training-free structured pruning method that trims multilayer perceptron layers while preserving attention mechanisms for reasoning tasks. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 15:06	Not relevant Proposes a pruning method rather than a novel benchmark or evaluation technique.
Unlocking Structured Pruning for Reasoning Language Models via Attention-Preserving MLP Trimming The paper proposes a training-free structured pruning method that trims multilayer perceptron layers while preserving attention mechanisms for reasoning tasks. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 15:06	Not relevant Investigates model pruning and efficiency rather than deployment safety or jailbreak prevention.
Unlocking Structured Pruning for Reasoning Language Models via Attention-Preserving MLP Trimming The paper proposes a training-free structured pruning method that trims multilayer perceptron layers while preserving attention mechanisms for reasoning tasks. Preprint	LLMs for code via arxiv · 15 Oct, 15:06	Not relevant Addresses general mathematical reasoning rather than software engineering or code generation.
Unlocking Structured Pruning for Reasoning Language Models via Attention-Preserving MLP Trimming The paper proposes a training-free structured pruning method that trims multilayer perceptron layers while preserving attention mechanisms for reasoning tasks. Preprint	Multimodal models via arxiv · 15 Oct, 15:06	Not relevant Exclusively addresses text-based reasoning models without multimodal data.
Unlocking Structured Pruning for Reasoning Language Models via Attention-Preserving MLP Trimming The paper proposes a training-free structured pruning method that trims multilayer perceptron layers while preserving attention mechanisms for reasoning tasks. Preprint	AI agents via arxiv · 15 Oct, 15:06	Not relevant Focuses on MLP pruning for transformer layers, not multi-step agent architectures.
Unlocking Structured Pruning for Reasoning Language Models via Attention-Preserving MLP Trimming The paper proposes a training-free structured pruning method that trims multilayer perceptron layers while preserving attention mechanisms for reasoning tasks. Preprint	Retrieval-augmented generation (RAG) via arxiv · 15 Oct, 15:06	Not relevant Does not involve external document retrieval, grounding, or RAG pipelines.
Learning in Orbit: Onboard Training and Closed-Loop Attitude Control with a Thermal Segmentation Model on a Spacecraft The paper demonstrates onboard thermal segmentation training and closed-loop attitude control on a satellite in low Earth orbit. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 15:06	Not relevant Deals with embedded spacecraft fault tolerance instead of LLM safety or alignment.
Learning in Orbit: Onboard Training and Closed-Loop Attitude Control with a Thermal Segmentation Model on a Spacecraft The paper demonstrates onboard thermal segmentation training and closed-loop attitude control on a satellite in low Earth orbit. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 15:06	Not relevant Demonstrates an onboard robotics deployment rather than proposing LLM evaluation frameworks.
Learning in Orbit: Onboard Training and Closed-Loop Attitude Control with a Thermal Segmentation Model on a Spacecraft The paper demonstrates onboard thermal segmentation training and closed-loop attitude control on a satellite in low Earth orbit. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 15:06	Not relevant Applies to spacecraft attitude control rather than manufacturing inspection or factory defect detection.
Learning in Orbit: Onboard Training and Closed-Loop Attitude Control with a Thermal Segmentation Model on a Spacecraft The paper demonstrates onboard thermal segmentation training and closed-loop attitude control on a satellite in low Earth orbit. Preprint	AI agents via arxiv · 15 Oct, 15:06	Not relevant Involves satellite attitude control and onboard segmentation, not LLM agents or tool planning.
PairIT: Autoregressive Transformers for Low-Data Molecule Optimization The paper introduces a transformer framework trained on matched molecular pairs for prompt-guided, multi-property molecular optimization in low-data regimes. Preprint	AI agents via arxiv · 15 Oct, 15:06	Not relevant Focuses on molecular SMILES generation rather than multi-step agent planning or tool execution.
PairIT: Autoregressive Transformers for Low-Data Molecule Optimization The paper introduces a transformer framework trained on matched molecular pairs for prompt-guided, multi-property molecular optimization in low-data regimes. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 15:06	Not relevant Concerns biochemical property optimization rather than AI safety or system guardrails.
PairIT: Autoregressive Transformers for Low-Data Molecule Optimization The paper introduces a transformer framework trained on matched molecular pairs for prompt-guided, multi-property molecular optimization in low-data regimes. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 15:06	Not relevant Focuses on domain-specific molecule optimization rather than general LLM evaluation methodology.
PairIT: Autoregressive Transformers for Low-Data Molecule Optimization The paper introduces a transformer framework trained on matched molecular pairs for prompt-guided, multi-property molecular optimization in low-data regimes. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address model compression, efficient serving, or small language models.
Imagining Goals with Geometric Horizon Models for Policy Generalization The paper proposes sampling from a geometric horizon world model to imagine novel outcomes for zero-shot policy generalization. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 15:06	Not relevant Evaluates RL policy generalization on existing benchmarks rather than proposing LLM benchmarks.
Imagining Goals with Geometric Horizon Models for Policy Generalization The paper proposes sampling from a geometric horizon world model to imagine novel outcomes for zero-shot policy generalization. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 15:06	Not relevant Does not study safety, guardrails, hallucinations, or deployment reliability.
Imagining Goals with Geometric Horizon Models for Policy Generalization The paper proposes sampling from a geometric horizon world model to imagine novel outcomes for zero-shot policy generalization. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Not about model compression, quantization, distillation, or efficient inference.
Imagining Goals with Geometric Horizon Models for Policy Generalization The paper proposes sampling from a geometric horizon world model to imagine novel outcomes for zero-shot policy generalization. Preprint	AI agents via arxiv · 15 Oct, 15:06	Not relevant Focuses on reinforcement learning policy generalization, not LLM planning or tool use.
Plant Memory: Technical Language Processing and a Verified Maintenance Ontology for Agent-Accessible Failure Knowledge from SAP PM Work Orders The paper presents a system using technical language processing to transform industrial maintenance work order records into an ontology. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 15:06	Not relevant Focuses on ontology building rather than general LLM evaluation benchmarks or methods.
Plant Memory: Technical Language Processing and a Verified Maintenance Ontology for Agent-Accessible Failure Knowledge from SAP PM Work Orders The paper presents a system using technical language processing to transform industrial maintenance work order records into an ontology. Preprint	AI agents via arxiv · 15 Oct, 15:06	Not relevant Describes ontology construction from maintenance logs rather than autonomous planning agents.
Plant Memory: Technical Language Processing and a Verified Maintenance Ontology for Agent-Accessible Failure Knowledge from SAP PM Work Orders The paper presents a system using technical language processing to transform industrial maintenance work order records into an ontology. Preprint	LLMs for code via arxiv · 15 Oct, 15:06	Not relevant Applies NLP to industrial maintenance logs rather than code generation or repair.
Plant Memory: Technical Language Processing and a Verified Maintenance Ontology for Agent-Accessible Failure Knowledge from SAP PM Work Orders The paper presents a system using technical language processing to transform industrial maintenance work order records into an ontology. Preprint	Multimodal models via arxiv · 15 Oct, 15:06	Not relevant Processes industrial free-text maintenance orders without multimodal image or audio modalities.
Plant Memory: Technical Language Processing and a Verified Maintenance Ontology for Agent-Accessible Failure Knowledge from SAP PM Work Orders The paper presents a system using technical language processing to transform industrial maintenance work order records into an ontology. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address small language models, quantization, pruning, or serving efficiency.

7,261–7,320 of 17,313← Previous Next →