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
Beyond Reachability: A Relation-Semantic Reasoning Framework for Biomedical Multi-hop QA The paper proposes a relation-semantic reasoning framework to resolve disconnects between answer reachability and valid paths in biomedical graphs. Preprint	AI agents via arxiv · 15 Oct, 15:06	Not relevant Focuses on biomedical graph multi-hop QA, not AI agents.
Beyond Reachability: A Relation-Semantic Reasoning Framework for Biomedical Multi-hop QA The paper proposes a relation-semantic reasoning framework to resolve disconnects between answer reachability and valid paths in biomedical graphs. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address model compression or inference efficiency.
Beyond Reachability: A Relation-Semantic Reasoning Framework for Biomedical Multi-hop QA The paper proposes a relation-semantic reasoning framework to resolve disconnects between answer reachability and valid paths in biomedical graphs. Preprint	Multimodal models via arxiv · 15 Oct, 15:06	Not relevant Knowledge graphs and text, not multimodal models.
Qwen-MT-Lumen at WMT2026: Instruction-Specialized Translation via Quality-Aware Training and On-Policy Distillation The paper presents a four-stage training pipeline combining preference optimization and distillation for instruction-following machine translation. Preprint	Retrieval-augmented generation (RAG) via arxiv · 15 Oct, 15:06	Not relevant Does not focus on retrieval-augmented generation.
Qwen-MT-Lumen at WMT2026: Instruction-Specialized Translation via Quality-Aware Training and On-Policy Distillation The paper presents a four-stage training pipeline combining preference optimization and distillation for instruction-following machine translation. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 15:06	Not relevant Instruction-following translation, not AI safety or guardrails.
Qwen-MT-Lumen at WMT2026: Instruction-Specialized Translation via Quality-Aware Training and On-Policy Distillation The paper presents a four-stage training pipeline combining preference optimization and distillation for instruction-following machine translation. Preprint	Multimodal models via arxiv · 15 Oct, 15:06	Not relevant Text translation only, not multimodal.
Qwen-MT-Lumen at WMT2026: Instruction-Specialized Translation via Quality-Aware Training and On-Policy Distillation The paper presents a four-stage training pipeline combining preference optimization and distillation for instruction-following machine translation. Preprint	LLMs for code via arxiv · 15 Oct, 15:06	Not relevant Not related to software coding.
Qwen-MT-Lumen at WMT2026: Instruction-Specialized Translation via Quality-Aware Training and On-Policy Distillation The paper presents a four-stage training pipeline combining preference optimization and distillation for instruction-following machine translation. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 15:06	Not relevant System description for a shared task, not a general evaluation benchmark.
Qwen-MT-Lumen at WMT2026: Instruction-Specialized Translation via Quality-Aware Training and On-Policy Distillation The paper presents a four-stage training pipeline combining preference optimization and distillation for instruction-following machine translation. Preprint	AI agents via arxiv · 15 Oct, 15:06	Not relevant Focuses on machine translation fine-tuning, not agents.
Likelihood Is Not Authorship: Can We Reliably Measure Human Contribution in AI-Assisted Writing? The paper reveals vulnerabilities in prompt-predictability metrics used to estimate human contribution in AI-assisted writing. Preprint	Retrieval-augmented generation (RAG) via arxiv · 15 Oct, 15:06	Not relevant Does not address retrieval-augmented generation.
Likelihood Is Not Authorship: Can We Reliably Measure Human Contribution in AI-Assisted Writing? The paper reveals vulnerabilities in prompt-predictability metrics used to estimate human contribution in AI-assisted writing. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address model efficiency or compression.
Likelihood Is Not Authorship: Can We Reliably Measure Human Contribution in AI-Assisted Writing? The paper reveals vulnerabilities in prompt-predictability metrics used to estimate human contribution in AI-assisted writing. Preprint	LLMs for code via arxiv · 15 Oct, 15:06	Not relevant Focuses on text authorship attribution, not code generation.
Likelihood Is Not Authorship: Can We Reliably Measure Human Contribution in AI-Assisted Writing? The paper reveals vulnerabilities in prompt-predictability metrics used to estimate human contribution in AI-assisted writing. Preprint	Multimodal models via arxiv · 15 Oct, 15:06	Not relevant Text attribution, not multimodal models.
Likelihood Is Not Authorship: Can We Reliably Measure Human Contribution in AI-Assisted Writing? The paper reveals vulnerabilities in prompt-predictability metrics used to estimate human contribution in AI-assisted writing. Preprint	AI agents via arxiv · 15 Oct, 15:06	Not relevant Studies human authorship estimation in AI text, not agents.
More Methods, Better Agents? A Reality Check on Self-Distillation in Agentic Reinforcement Learning The paper benchmarks ten self-distillation methods in agentic reinforcement learning across varied model sizes, tasks, and computational costs. Preprint	Retrieval-augmented generation (RAG) via arxiv · 15 Oct, 15:06	Not relevant Does not address retrieval-augmented generation.
More Methods, Better Agents? A Reality Check on Self-Distillation in Agentic Reinforcement Learning The paper benchmarks ten self-distillation methods in agentic reinforcement learning across varied model sizes, tasks, and computational costs. Preprint	Multimodal models via arxiv · 15 Oct, 15:06	Not relevant Text-based agent environments, not multimodal models.
More Methods, Better Agents? A Reality Check on Self-Distillation in Agentic Reinforcement Learning The paper benchmarks ten self-distillation methods in agentic reinforcement learning across varied model sizes, tasks, and computational costs. Preprint	LLMs for code via arxiv · 15 Oct, 15:06	Not relevant Evaluates web/science agent environments, not code generation.
More Methods, Better Agents? A Reality Check on Self-Distillation in Agentic Reinforcement Learning The paper benchmarks ten self-distillation methods in agentic reinforcement learning across varied model sizes, tasks, and computational costs. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 15:06	Not relevant Focuses on training stability and performance, not safety or guardrails.
JetSpec++: Foresight Parallel Drafting with Self Refinements The paper introduces a speculative decoding system combining self-refinement and pipelined drafting to reduce wall-clock generation time. Preprint	Retrieval-augmented generation (RAG) via arxiv · 15 Oct, 15:06	Not relevant Does not involve retrieval-augmented generation.
JetSpec++: Foresight Parallel Drafting with Self Refinements The paper introduces a speculative decoding system combining self-refinement and pipelined drafting to reduce wall-clock generation time. Preprint	AI agents via arxiv · 15 Oct, 15:06	Not relevant Focuses on speculative decoding, not AI agents.
JetSpec++: Foresight Parallel Drafting with Self Refinements The paper introduces a speculative decoding system combining self-refinement and pipelined drafting to reduce wall-clock generation time. Preprint	LLMs for code via arxiv · 15 Oct, 15:06	Not relevant Not focused on code generation or software engineering.
JetSpec++: Foresight Parallel Drafting with Self Refinements The paper introduces a speculative decoding system combining self-refinement and pipelined drafting to reduce wall-clock generation time. Preprint	Multimodal models via arxiv · 15 Oct, 15:06	Not relevant Applies to autoregressive language models, not multimodal.
JetSpec++: Foresight Parallel Drafting with Self Refinements The paper introduces a speculative decoding system combining self-refinement and pipelined drafting to reduce wall-clock generation time. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 15:06	Not relevant Introduces an inference acceleration method, not an evaluation benchmark.
JetSpec++: Foresight Parallel Drafting with Self Refinements The paper introduces a speculative decoding system combining self-refinement and pipelined drafting to reduce wall-clock generation time. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 15:06	Not relevant Does not address AI safety or guardrails.
Evolve Together, Fall into Place: How Social Scaffolding Shapes Self-Evolving Multi-Agent Systems The paper investigates how team composition and organizational interdependence affect performance in self-evolving multi-agent systems. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address model compression or serving efficiency.
Evolve Together, Fall into Place: How Social Scaffolding Shapes Self-Evolving Multi-Agent Systems The paper investigates how team composition and organizational interdependence affect performance in self-evolving multi-agent systems. Preprint	LLMs for code via arxiv · 15 Oct, 15:06	Not relevant Codebase understanding is only one test task; core focus is multi-agent systems.
Evolve Together, Fall into Place: How Social Scaffolding Shapes Self-Evolving Multi-Agent Systems The paper investigates how team composition and organizational interdependence affect performance in self-evolving multi-agent systems. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 15:06	Not relevant Focuses on multi-agent collaboration rather than AI safety or reliability.
Evolve Together, Fall into Place: How Social Scaffolding Shapes Self-Evolving Multi-Agent Systems The paper investigates how team composition and organizational interdependence affect performance in self-evolving multi-agent systems. Preprint	Retrieval-augmented generation (RAG) via arxiv · 15 Oct, 15:06	Not relevant Focuses on multi-agent collaboration structures, not RAG systems.
Evolve Together, Fall into Place: How Social Scaffolding Shapes Self-Evolving Multi-Agent Systems The paper investigates how team composition and organizational interdependence affect performance in self-evolving multi-agent systems. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 15:06	Not relevant Evaluates multi-agent evolution, not benchmark design.
Evolve Together, Fall into Place: How Social Scaffolding Shapes Self-Evolving Multi-Agent Systems The paper investigates how team composition and organizational interdependence affect performance in self-evolving multi-agent systems. Preprint	Multimodal models via arxiv · 15 Oct, 15:06	Not relevant Does not involve multimodal models.
Delta Tokenizer: Reference-Identical Tokenization of Session Deltas and Segment-Parallelism for Long Prompts on Multicore CPUs The paper introduces a CPU front-end tokenizer that encodes only session deltas to speed up multi-turn inference. Preprint	Retrieval-augmented generation (RAG) via arxiv · 15 Oct, 15:06	Not relevant Does not involve retrieval-augmented generation.
Delta Tokenizer: Reference-Identical Tokenization of Session Deltas and Segment-Parallelism for Long Prompts on Multicore CPUs The paper introduces a CPU front-end tokenizer that encodes only session deltas to speed up multi-turn inference. Preprint	AI agents via arxiv · 15 Oct, 15:06	Not relevant Optimizes CPU tokenization for agent sessions, not agent planning or architecture.
Delta Tokenizer: Reference-Identical Tokenization of Session Deltas and Segment-Parallelism for Long Prompts on Multicore CPUs The paper introduces a CPU front-end tokenizer that encodes only session deltas to speed up multi-turn inference. Preprint	LLMs for code via arxiv · 15 Oct, 15:06	Not relevant Focuses on tokenizer serving systems, not code LLMs.
Delta Tokenizer: Reference-Identical Tokenization of Session Deltas and Segment-Parallelism for Long Prompts on Multicore CPUs The paper introduces a CPU front-end tokenizer that encodes only session deltas to speed up multi-turn inference. Preprint	Multimodal models via arxiv · 15 Oct, 15:06	Not relevant Text tokenization, not multimodal.
Delta Tokenizer: Reference-Identical Tokenization of Session Deltas and Segment-Parallelism for Long Prompts on Multicore CPUs The paper introduces a CPU front-end tokenizer that encodes only session deltas to speed up multi-turn inference. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 15:06	Not relevant Systems paper on serving speed, not LLM evaluation benchmarks.
Delta Tokenizer: Reference-Identical Tokenization of Session Deltas and Segment-Parallelism for Long Prompts on Multicore CPUs The paper introduces a CPU front-end tokenizer that encodes only session deltas to speed up multi-turn inference. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 15:06	Not relevant Does not address safety, hallucination, or alignment.
Model vs. User: Whose Values Prevail? Inference-Time Value Alignment in Multi-Turn Conversations The paper studies the durability of inference-time value steering in multi-turn conversations when user and model values conflict. Preprint	LLMs for code via arxiv · 15 Oct, 15:06	Not relevant Not related to code generation.
Model vs. User: Whose Values Prevail? Inference-Time Value Alignment in Multi-Turn Conversations The paper studies the durability of inference-time value steering in multi-turn conversations when user and model values conflict. Preprint	Multimodal models via arxiv · 15 Oct, 15:06	Not relevant Text-based conversational alignment, not multimodal.
Model vs. User: Whose Values Prevail? Inference-Time Value Alignment in Multi-Turn Conversations The paper studies the durability of inference-time value steering in multi-turn conversations when user and model values conflict. Preprint	Retrieval-augmented generation (RAG) via arxiv · 15 Oct, 15:06	Not relevant Does not address retrieval-augmented generation.
Model vs. User: Whose Values Prevail? Inference-Time Value Alignment in Multi-Turn Conversations The paper studies the durability of inference-time value steering in multi-turn conversations when user and model values conflict. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not focus on efficient serving or small models.
Model vs. User: Whose Values Prevail? Inference-Time Value Alignment in Multi-Turn Conversations The paper studies the durability of inference-time value steering in multi-turn conversations when user and model values conflict. Preprint	AI agents via arxiv · 15 Oct, 15:06	Not relevant Evaluates multi-turn value steering, not agent planning or tool use.
MAPLE: A Multi-Axis Parametric Knowledge Extraction Lens for LLM Evaluation The paper presents a diagnostic evaluation framework to measure how inference-time techniques affect parametric knowledge extraction in models. Preprint	Retrieval-augmented generation (RAG) via arxiv · 15 Oct, 15:06	Not relevant Studies parametric internal knowledge extraction, not external retrieval.
MAPLE: A Multi-Axis Parametric Knowledge Extraction Lens for LLM Evaluation The paper presents a diagnostic evaluation framework to measure how inference-time techniques affect parametric knowledge extraction in models. Preprint	Multimodal models via arxiv · 15 Oct, 15:06	Not relevant Text-only evaluation, not multimodal.
MAPLE: A Multi-Axis Parametric Knowledge Extraction Lens for LLM Evaluation The paper presents a diagnostic evaluation framework to measure how inference-time techniques affect parametric knowledge extraction in models. Preprint	LLMs for code via arxiv · 15 Oct, 15:06	Not relevant Not focused on code or programming.
MAPLE: A Multi-Axis Parametric Knowledge Extraction Lens for LLM Evaluation The paper presents a diagnostic evaluation framework to measure how inference-time techniques affect parametric knowledge extraction in models. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 15:06	Not relevant Focuses on diagnostic extraction metrics, not safety guardrails or jailbreaks.
MAPLE: A Multi-Axis Parametric Knowledge Extraction Lens for LLM Evaluation The paper presents a diagnostic evaluation framework to measure how inference-time techniques affect parametric knowledge extraction in models. Preprint	AI agents via arxiv · 15 Oct, 15:06	Not relevant Does not focus on AI agents or tool planning.
MAPLE: A Multi-Axis Parametric Knowledge Extraction Lens for LLM Evaluation The paper presents a diagnostic evaluation framework to measure how inference-time techniques affect parametric knowledge extraction in models. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Studies knowledge extraction techniques, not model compression or efficiency.
EGCForest: Structured Agent Memory Organization for Enterprise Group Chats The paper proposes a multi-view hierarchical memory framework to improve information retrieval from enterprise group chats. Preprint	Multimodal models via arxiv · 15 Oct, 15:06	Not relevant Text-based chat retrieval, not multimodal.
EGCForest: Structured Agent Memory Organization for Enterprise Group Chats The paper proposes a multi-view hierarchical memory framework to improve information retrieval from enterprise group chats. Preprint	LLMs for code via arxiv · 15 Oct, 15:06	Not relevant Focuses on enterprise chat retrieval, not code generation.
EGCForest: Structured Agent Memory Organization for Enterprise Group Chats The paper proposes a multi-view hierarchical memory framework to improve information retrieval from enterprise group chats. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 15:06	Not relevant Proposes a retrieval framework, not general benchmark methodology.
EGCForest: Structured Agent Memory Organization for Enterprise Group Chats The paper proposes a multi-view hierarchical memory framework to improve information retrieval from enterprise group chats. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not focus on model compression or inference efficiency.
EGCForest: Structured Agent Memory Organization for Enterprise Group Chats The paper proposes a multi-view hierarchical memory framework to improve information retrieval from enterprise group chats. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 15:06	Not relevant Focuses on retrieval accuracy, not AI safety or guardrails.
OpenTWBench (Open Taiwan Benchmark): Taiwan-Domain Knowledge Evaluation for Large Language Models, and How the Measurement Apparatus Shapes the Score The paper introduces a two-part benchmark to evaluate language models on general and examination knowledge concerning Taiwan. Preprint	LLMs for code via arxiv · 15 Oct, 15:06	Not relevant Not about coding LLMs.
OpenTWBench (Open Taiwan Benchmark): Taiwan-Domain Knowledge Evaluation for Large Language Models, and How the Measurement Apparatus Shapes the Score The paper introduces a two-part benchmark to evaluate language models on general and examination knowledge concerning Taiwan. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 15:06	Not relevant Focuses on domain knowledge testing, not AI safety or robustness.
OpenTWBench (Open Taiwan Benchmark): Taiwan-Domain Knowledge Evaluation for Large Language Models, and How the Measurement Apparatus Shapes the Score The paper introduces a two-part benchmark to evaluate language models on general and examination knowledge concerning Taiwan. Preprint	Retrieval-augmented generation (RAG) via arxiv · 15 Oct, 15:06	Not relevant Does not address retrieval-augmented generation.
OpenTWBench (Open Taiwan Benchmark): Taiwan-Domain Knowledge Evaluation for Large Language Models, and How the Measurement Apparatus Shapes the Score The paper introduces a two-part benchmark to evaluate language models on general and examination knowledge concerning Taiwan. Preprint	AI agents via arxiv · 15 Oct, 15:06	Not relevant Does not address AI agents.
OpenTWBench (Open Taiwan Benchmark): Taiwan-Domain Knowledge Evaluation for Large Language Models, and How the Measurement Apparatus Shapes the Score The paper introduces a two-part benchmark to evaluate language models on general and examination knowledge concerning Taiwan. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not focus on model efficiency or compression.
OpenTWBench (Open Taiwan Benchmark): Taiwan-Domain Knowledge Evaluation for Large Language Models, and How the Measurement Apparatus Shapes the Score The paper introduces a two-part benchmark to evaluate language models on general and examination knowledge concerning Taiwan. Preprint	Multimodal models via arxiv · 15 Oct, 15:06	Not relevant Focuses on text-based knowledge benchmarks, not multimodal models.
SEA-HELM Audio: Evaluating Multilingual ASR-LLM Pipeline Performance Across SEA-HELM Subsets The paper presents an audio evaluation benchmark assessing accuracy drops between speech-input and text-input pipelines in Southeast Asian languages. Preprint	Small and efficient models via arxiv · 15 Oct, 15:06	Not relevant Does not address model compression or efficient serving.
SEA-HELM Audio: Evaluating Multilingual ASR-LLM Pipeline Performance Across SEA-HELM Subsets The paper presents an audio evaluation benchmark assessing accuracy drops between speech-input and text-input pipelines in Southeast Asian languages. Preprint	LLMs for code via arxiv · 15 Oct, 15:06	Not relevant Not related to code generation.

3,421–3,480 of 13,754← Previous Next →