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
RAMP: Recurrent Adaptive Mixed Precision for Efficient Prefill in Recurrent-Depth Language Models Analyzes quantization error propagation in recurrent-depth language models and proposes adaptive mixed-precision schedules for faster prefill inference. Preprint	AI agents via arxiv · 12 Oct, 16:28	Not relevant Deals with recurrent-depth model quantization, not agent orchestration.
CliQ-RRG: Clinical-Knowledge Guided Disease-aware Visual-Textual Alignment for QA-Style Radiology Report Generation Presents a two-stage visual-textual alignment and retrieval-augmented framework for generating structured radiology reports from medical scans. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:28	Not relevant Targets medical report generation quality rather than AI safety or guardrails.
CliQ-RRG: Clinical-Knowledge Guided Disease-aware Visual-Textual Alignment for QA-Style Radiology Report Generation Presents a two-stage visual-textual alignment and retrieval-augmented framework for generating structured radiology reports from medical scans. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:28	Not relevant Proposes a domain-specific model framework rather than an evaluation benchmark.
CliQ-RRG: Clinical-Knowledge Guided Disease-aware Visual-Textual Alignment for QA-Style Radiology Report Generation Presents a two-stage visual-textual alignment and retrieval-augmented framework for generating structured radiology reports from medical scans. Preprint	AI agents via arxiv · 12 Oct, 16:28	Not relevant Presents a report generation pipeline rather than multi-step autonomous planning agents.
CliQ-RRG: Clinical-Knowledge Guided Disease-aware Visual-Textual Alignment for QA-Style Radiology Report Generation Presents a two-stage visual-textual alignment and retrieval-augmented framework for generating structured radiology reports from medical scans. Preprint	LLMs for code via arxiv · 12 Oct, 16:28	Not relevant Focuses on medical radiology reporting, not code development.
CliQ-RRG: Clinical-Knowledge Guided Disease-aware Visual-Textual Alignment for QA-Style Radiology Report Generation Presents a two-stage visual-textual alignment and retrieval-augmented framework for generating structured radiology reports from medical scans. Preprint	Small and efficient models via arxiv · 12 Oct, 16:28	Not relevant Does not focus on model compression or inference efficiency.
KV-Repurpose: A Testbed for Evaluating Non-Prefix KV-Cache Reuse Develops an open-source testbed to evaluate segment-wise key-value cache reuse beyond exact prompt prefixes during language model serving. Preprint	AI agents via arxiv · 12 Oct, 16:28	Not relevant Focuses on serving-level KV-cache memory management rather than multi-step agent planning.
KV-Repurpose: A Testbed for Evaluating Non-Prefix KV-Cache Reuse Develops an open-source testbed to evaluate segment-wise key-value cache reuse beyond exact prompt prefixes during language model serving. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:28	Not relevant Proposes a serving system testbed for cache hits, not a standard LLM evaluation benchmark.
KV-Repurpose: A Testbed for Evaluating Non-Prefix KV-Cache Reuse Develops an open-source testbed to evaluate segment-wise key-value cache reuse beyond exact prompt prefixes during language model serving. Preprint	LLMs for code via arxiv · 12 Oct, 16:28	Not relevant Targets serving systems, not code generation or program repair.
KV-Repurpose: A Testbed for Evaluating Non-Prefix KV-Cache Reuse Develops an open-source testbed to evaluate segment-wise key-value cache reuse beyond exact prompt prefixes during language model serving. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:28	Not relevant Addresses latency and caching efficiency, not AI safety or alignment.
KV-Repurpose: A Testbed for Evaluating Non-Prefix KV-Cache Reuse Develops an open-source testbed to evaluate segment-wise key-value cache reuse beyond exact prompt prefixes during language model serving. Preprint	Multimodal models via arxiv · 12 Oct, 16:28	Not relevant Focuses on text LLM serving infrastructure without multimodal data.
From 2.1M Hours of Raw Broadcast Audio to Finnish Speech SSL: Confidence Filtering and Architectural Analysis at Scale Evaluates speech self-supervised learning models on Finnish broadcast audio using ASR confidence filtering and controlled architectural comparisons. Preprint	LLMs for code via arxiv · 12 Oct, 16:28	Not relevant Concerned with speech audio representation, not code generation.
From 2.1M Hours of Raw Broadcast Audio to Finnish Speech SSL: Confidence Filtering and Architectural Analysis at Scale Evaluates speech self-supervised learning models on Finnish broadcast audio using ASR confidence filtering and controlled architectural comparisons. Preprint	Multimodal models via arxiv · 12 Oct, 16:28	Not relevant Focuses exclusively on audio/speech SSL rather than joint multimodal vision-language models.
From 2.1M Hours of Raw Broadcast Audio to Finnish Speech SSL: Confidence Filtering and Architectural Analysis at Scale Evaluates speech self-supervised learning models on Finnish broadcast audio using ASR confidence filtering and controlled architectural comparisons. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:28	Not relevant Deals with speech SSL data quality, not deployment safety or jailbreaks.
From 2.1M Hours of Raw Broadcast Audio to Finnish Speech SSL: Confidence Filtering and Architectural Analysis at Scale Evaluates speech self-supervised learning models on Finnish broadcast audio using ASR confidence filtering and controlled architectural comparisons. Preprint	Small and efficient models via arxiv · 12 Oct, 16:28	Not relevant Studies speech SSL data filtering and architecture comparisons, not model compression.
From 2.1M Hours of Raw Broadcast Audio to Finnish Speech SSL: Confidence Filtering and Architectural Analysis at Scale Evaluates speech self-supervised learning models on Finnish broadcast audio using ASR confidence filtering and controlled architectural comparisons. Preprint	Retrieval-augmented generation (RAG) via arxiv · 12 Oct, 16:28	Not relevant Does not investigate retrieval-augmented generation.
From 2.1M Hours of Raw Broadcast Audio to Finnish Speech SSL: Confidence Filtering and Architectural Analysis at Scale Evaluates speech self-supervised learning models on Finnish broadcast audio using ASR confidence filtering and controlled architectural comparisons. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:28	Not relevant Presents an audio pretraining filtering method, not an LLM benchmark.
From 2.1M Hours of Raw Broadcast Audio to Finnish Speech SSL: Confidence Filtering and Architectural Analysis at Scale Evaluates speech self-supervised learning models on Finnish broadcast audio using ASR confidence filtering and controlled architectural comparisons. Preprint	AI agents via arxiv · 12 Oct, 16:28	Not relevant Focuses on Finnish speech self-supervised learning, not autonomous agents.
BenCoDiC: A Large-Scale Contextual Cyberbullying Corpus for Bengali Political News Discourse FULL VERSION Presents a multi-dimensional annotated corpus of Bengali political YouTube discussions to support research on contextual cyberbullying detection. Preprint	AI agents via arxiv · 12 Oct, 16:28	Not relevant Presents a cyberbullying text corpus rather than an autonomous LLM agent architecture.
BenCoDiC: A Large-Scale Contextual Cyberbullying Corpus for Bengali Political News Discourse FULL VERSION Presents a multi-dimensional annotated corpus of Bengali political YouTube discussions to support research on contextual cyberbullying detection. Preprint	Retrieval-augmented generation (RAG) via arxiv · 12 Oct, 16:28	Not relevant Focuses on social media discourse dataset creation, not RAG systems.
BenCoDiC: A Large-Scale Contextual Cyberbullying Corpus for Bengali Political News Discourse FULL VERSION Presents a multi-dimensional annotated corpus of Bengali political YouTube discussions to support research on contextual cyberbullying detection. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:28	Not relevant Presents a detection dataset rather than LLM safety, red-teaming, or guardrails.
BenCoDiC: A Large-Scale Contextual Cyberbullying Corpus for Bengali Political News Discourse FULL VERSION Presents a multi-dimensional annotated corpus of Bengali political YouTube discussions to support research on contextual cyberbullying detection. Preprint	LLMs for code via arxiv · 12 Oct, 16:28	Not relevant Concerned with political social media text, not coding or software development.
BenCoDiC: A Large-Scale Contextual Cyberbullying Corpus for Bengali Political News Discourse FULL VERSION Presents a multi-dimensional annotated corpus of Bengali political YouTube discussions to support research on contextual cyberbullying detection. Preprint	Small and efficient models via arxiv · 12 Oct, 16:28	Not relevant Does not explore model compression, distillation, or efficient inference.
BenCoDiC: A Large-Scale Contextual Cyberbullying Corpus for Bengali Political News Discourse FULL VERSION Presents a multi-dimensional annotated corpus of Bengali political YouTube discussions to support research on contextual cyberbullying detection. Preprint	Multimodal models via arxiv · 12 Oct, 16:28	Not relevant Deals with textual commentary from YouTube, not multimodal model architectures.
BenCoDiC: A Large-Scale Contextual Cyberbullying Corpus for Bengali Political News Discourse FULL VERSION Presents a multi-dimensional annotated corpus of Bengali political YouTube discussions to support research on contextual cyberbullying detection. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:28	Not relevant Introduces an annotated dataset for cyberbullying detection, not general LLM benchmarking.
Neurograf v1: An End-to-End Romanian-First Language Model Family Introduces an end-to-end open pipeline, pretraining corpus, and family of decoder-only models trained from scratch for Romanian. Preprint	LLMs for code via arxiv · 12 Oct, 16:28	Not relevant Focuses on Romanian general text models, not software engineering or coding.
Neurograf v1: An End-to-End Romanian-First Language Model Family Introduces an end-to-end open pipeline, pretraining corpus, and family of decoder-only models trained from scratch for Romanian. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:28	Not relevant Focuses on low-resource language pretraining rather than safety or reliability.
Neurograf v1: An End-to-End Romanian-First Language Model Family Introduces an end-to-end open pipeline, pretraining corpus, and family of decoder-only models trained from scratch for Romanian. Preprint	Retrieval-augmented generation (RAG) via arxiv · 12 Oct, 16:28	Not relevant Presents a pretraining pipeline rather than retrieval-augmented generation.
Neurograf v1: An End-to-End Romanian-First Language Model Family Introduces an end-to-end open pipeline, pretraining corpus, and family of decoder-only models trained from scratch for Romanian. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:28	Not relevant Introduces a model family and dataset rather than a novel LLM evaluation methodology.
Neurograf v1: An End-to-End Romanian-First Language Model Family Introduces an end-to-end open pipeline, pretraining corpus, and family of decoder-only models trained from scratch for Romanian. Preprint	AI agents via arxiv · 12 Oct, 16:28	Not relevant Introduces a pretrained Romanian language model family, not agent planning or tool use.
Neurograf v1: An End-to-End Romanian-First Language Model Family Introduces an end-to-end open pipeline, pretraining corpus, and family of decoder-only models trained from scratch for Romanian. Preprint	Multimodal models via arxiv · 12 Oct, 16:28	Not relevant The models are unimodal text-based language models.
Les Misérables Dialogue: A film-based network representation of social communication Constructs a temporal and multilayer social dialogue network dataset based on the film adaptation of Les Misérables. Preprint	Multimodal models via arxiv · 12 Oct, 16:28	Not relevant Focuses on text dialogue networks rather than multimodal foundation models.
Les Misérables Dialogue: A film-based network representation of social communication Constructs a temporal and multilayer social dialogue network dataset based on the film adaptation of Les Misérables. Preprint	AI agents via arxiv · 12 Oct, 16:28	Not relevant Constructs a temporal text network from a movie, not an AI agent system.
Les Misérables Dialogue: A film-based network representation of social communication Constructs a temporal and multilayer social dialogue network dataset based on the film adaptation of Les Misérables. Preprint	LLMs for code via arxiv · 12 Oct, 16:28	Not relevant Focuses on narrative dialogue networks, not code generation or repair.
Les Misérables Dialogue: A film-based network representation of social communication Constructs a temporal and multilayer social dialogue network dataset based on the film adaptation of Les Misérables. Preprint	Retrieval-augmented generation (RAG) via arxiv · 12 Oct, 16:28	Not relevant Deals with dialogue network analysis, not retrieval-augmented generation.
Les Misérables Dialogue: A film-based network representation of social communication Constructs a temporal and multilayer social dialogue network dataset based on the film adaptation of Les Misérables. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:28	Not relevant Addresses social dialogue graphs rather than AI deployment safety or jailbreaks.
Les Misérables Dialogue: A film-based network representation of social communication Constructs a temporal and multilayer social dialogue network dataset based on the film adaptation of Les Misérables. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:28	Not relevant Presents a social network dataset from a movie, not an LLM benchmark.
Les Misérables Dialogue: A film-based network representation of social communication Constructs a temporal and multilayer social dialogue network dataset based on the film adaptation of Les Misérables. Preprint	Small and efficient models via arxiv · 12 Oct, 16:28	Not relevant Not related to model efficiency, quantization, or serving.
SOAR: Source-Aware Reinforcement Learning with Precise Credit Assignment for Search Agents Proposes a reinforcement learning framework that tracks external versus internal knowledge sources to reduce search agent fabrication. Preprint	Multimodal models via arxiv · 12 Oct, 16:28	Not relevant Centers on text-based search reasoning without multimodal inputs.
SOAR: Source-Aware Reinforcement Learning with Precise Credit Assignment for Search Agents Proposes a reinforcement learning framework that tracks external versus internal knowledge sources to reduce search agent fabrication. Preprint	Small and efficient models via arxiv · 12 Oct, 16:28	Not relevant Focuses on credit assignment and grounding, not inference acceleration or quantization.
SOAR: Source-Aware Reinforcement Learning with Precise Credit Assignment for Search Agents Proposes a reinforcement learning framework that tracks external versus internal knowledge sources to reduce search agent fabrication. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:28	Not relevant Proposes an RL training framework rather than an evaluation benchmark.
SOAR: Source-Aware Reinforcement Learning with Precise Credit Assignment for Search Agents Proposes a reinforcement learning framework that tracks external versus internal knowledge sources to reduce search agent fabrication. Preprint	LLMs for code via arxiv · 12 Oct, 16:28	Not relevant Focuses on general web search and question answering, not programming or code tasks.
Scaling Agentic Data for Long-Horizon Terminal Intelligence Introduces a synthetic data generation framework to produce long-horizon task trajectories for terminal agents. Preprint	Multimodal models via arxiv · 12 Oct, 16:28	Not relevant Operates on terminal text commands without vision or multimodal input.
Scaling Agentic Data for Long-Horizon Terminal Intelligence Introduces a synthetic data generation framework to produce long-horizon task trajectories for terminal agents. Preprint	Small and efficient models via arxiv · 12 Oct, 16:28	Not relevant Focuses on agentic training trajectories rather than model compression or inference efficiency.
Scaling Agentic Data for Long-Horizon Terminal Intelligence Introduces a synthetic data generation framework to produce long-horizon task trajectories for terminal agents. Preprint	Retrieval-augmented generation (RAG) via arxiv · 12 Oct, 16:28	Not relevant Focuses on terminal agent trajectories rather than retrieval-augmented generation.
Scaling Agentic Data for Long-Horizon Terminal Intelligence Introduces a synthetic data generation framework to produce long-horizon task trajectories for terminal agents. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:28	Not relevant Proposes a data synthesis engine for training, not an evaluation benchmark.
Scaling Agentic Data for Long-Horizon Terminal Intelligence Introduces a synthetic data generation framework to produce long-horizon task trajectories for terminal agents. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:28	Not relevant Focuses on long-horizon agent capabilities rather than safety guardrails or red-teaming.
PreSTO: Predictive Subtree Prefetching for Fast LLM Power Sampling Presents a predictive subtree prefetching method to accelerate power sampling for test-time language model reasoning. Preprint	Multimodal models via arxiv · 12 Oct, 16:28	Not relevant Addresses text LLM sampling without multimodal modalities.
PreSTO: Predictive Subtree Prefetching for Fast LLM Power Sampling Presents a predictive subtree prefetching method to accelerate power sampling for test-time language model reasoning. Preprint	AI agents via arxiv · 12 Oct, 16:28	Not relevant Focuses on test-time sampling efficiency rather than autonomous agent planning or tool use.
PreSTO: Predictive Subtree Prefetching for Fast LLM Power Sampling Presents a predictive subtree prefetching method to accelerate power sampling for test-time language model reasoning. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:28	Not relevant Focuses on sampling runtime acceleration, not safety, guardrails, or robustness.
PreSTO: Predictive Subtree Prefetching for Fast LLM Power Sampling Presents a predictive subtree prefetching method to accelerate power sampling for test-time language model reasoning. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:28	Not relevant Proposes an accelerated sampling method, not an evaluation benchmark or diagnostic.
PreSTO: Predictive Subtree Prefetching for Fast LLM Power Sampling Presents a predictive subtree prefetching method to accelerate power sampling for test-time language model reasoning. Preprint	Retrieval-augmented generation (RAG) via arxiv · 12 Oct, 16:28	Not relevant Does not involve external document retrieval or grounding.
PreSTO: Predictive Subtree Prefetching for Fast LLM Power Sampling Presents a predictive subtree prefetching method to accelerate power sampling for test-time language model reasoning. Preprint	LLMs for code via arxiv · 12 Oct, 16:28	Not relevant Applies to general reasoning sampling, not specifically to software engineering tasks.
S$^3$R: Fast and Faithful RL Rollouts via Self-Sparse Speculative Decoding Introduces speculative decoding using self-sparse attention to speed up rollout generation during reinforcement learning. Preprint	LLMs for code via arxiv · 12 Oct, 16:28	Not relevant Focuses on RL rollout efficiency in general, not code generation or program repair.
S$^3$R: Fast and Faithful RL Rollouts via Self-Sparse Speculative Decoding Introduces speculative decoding using self-sparse attention to speed up rollout generation during reinforcement learning. Preprint	AI safety and reliability in deployment via arxiv · 12 Oct, 16:28	Not relevant Focuses on rollout efficiency, not AI safety, alignment risks, or guardrails.
S$^3$R: Fast and Faithful RL Rollouts via Self-Sparse Speculative Decoding Introduces speculative decoding using self-sparse attention to speed up rollout generation during reinforcement learning. Preprint	LLM evaluation and benchmarks via arxiv · 12 Oct, 16:28	Not relevant Introduces a decoding method for RL, not an evaluation framework or benchmark.
S$^3$R: Fast and Faithful RL Rollouts via Self-Sparse Speculative Decoding Introduces speculative decoding using self-sparse attention to speed up rollout generation during reinforcement learning. Preprint	AI agents via arxiv · 12 Oct, 16:28	Not relevant Focuses on KV-cache optimization during RL training, not agent workflows or multi-step tool use.
S$^3$R: Fast and Faithful RL Rollouts via Self-Sparse Speculative Decoding Introduces speculative decoding using self-sparse attention to speed up rollout generation during reinforcement learning. Preprint	Multimodal models via arxiv · 12 Oct, 16:28	Not relevant Pertains strictly to language model attention decoding without multimodal data.
S$^3$R: Fast and Faithful RL Rollouts via Self-Sparse Speculative Decoding Introduces speculative decoding using self-sparse attention to speed up rollout generation during reinforcement learning. Preprint	Retrieval-augmented generation (RAG) via arxiv · 12 Oct, 16:28	Not relevant Deals with RL training optimization, not document retrieval or grounding.
SharP: Soft and Hard Prompt-Guided Augmentation with LLM for Low Resource Fake News Detection Proposes a prompt-guided text augmentation framework to assist fake news detection in low-resource settings. Peer-reviewed	Small and efficient models via arxiv · 12 Oct, 16:28	Not relevant Refers to low-data regimes, not small model compression or inference efficiency.

16,081–16,140 of 17,313← Previous Next →