Author

PaperAgent

Daily updates, analyzing cutting-edge AI research papers

216

Articles

Likes

412

Views

Comments

Latest from PaperAgent

100 recent articles max

PaperAgent

Apr 24, 2026 · Artificial Intelligence

DeepSeek‑V4 Open‑Sources Its Million‑Token Architecture and Calls Out Claude Opus 4.6

DeepSeek‑V4’s open‑source report reveals a hybrid CSA/HCA attention design, manifold‑constrained residuals and the Muon optimizer that cut per‑token FLOPs to 27 % and KV‑Cache to 10 % at 1 M tokens, while benchmark results show it outperforms Claude Opus 4.6 on most tasks yet still lags on complex instruction following and multi‑turn dialogue.

AI ArchitectureClaude OpusDeepSeek V4

0 likes · 11 min read

DeepSeek‑V4 Open‑Sources Its Million‑Token Architecture and Calls Out Claude Opus 4.6

PaperAgent

Apr 24, 2026 · Artificial Intelligence

Agent Skills Practical Guide: From Concept to Actionable AI Agents

The article explains Anthropic’s 2025 Agent Skills standard, how it enables AI to perform actions such as database queries and API calls, and provides a detailed guide covering its definition, modular design, industry adoption, and practical usage scenarios.

AI agentsAgent skillsAnthropic

0 likes · 3 min read

Agent Skills Practical Guide: From Concept to Actionable AI Agents

PaperAgent

Apr 23, 2026 · Artificial Intelligence

Stop RAG, Navigate Enterprise Knowledge Directly with CORPUS2SKILL

The article critiques traditional RAG’s blind spots, introduces CORPUS2SKILL’s offline‑compile, online‑navigate two‑stage architecture that builds a hierarchical topic tree and progressive‑disclosure skill files, and shows through WixQA benchmarks that this approach outperforms dense retrieval and Agentic RAG on F1, factuality and recall while highlighting cost and hierarchy quality trade‑offs.

Hierarchical ClusteringPrompt EngineeringRAG

0 likes · 7 min read

Stop RAG, Navigate Enterprise Knowledge Directly with CORPUS2SKILL

PaperAgent

Apr 22, 2026 · Artificial Intelligence

How SkillClaw Enables Collective Evolution of Agent Skills in Real-World Use

SkillClaw introduces a centralized evolution framework that transforms user interactions into structured evidence, allowing LLM agents to refine, create, or skip skills based on aggregated success and failure patterns, with nightly validation ensuring only proven improvements are deployed, resulting in consistent performance gains across diverse tasks.

AI workflowLLM AgentsSkill Evolution

0 likes · 13 min read

How SkillClaw Enables Collective Evolution of Agent Skills in Real-World Use

PaperAgent

Apr 22, 2026 · Artificial Intelligence

Alibaba Unveils Four New Open‑Source Qwen3.6 Models: 27B Dense and 35B‑A3B MoE

Alibaba has added four new open‑source weight versions to its Qwen3.6 series, featuring the 27‑billion‑parameter dense multimodal model Qwen3.6‑27B and the 35‑billion‑parameter sparse expert model Qwen3.6‑35B‑A3B, both designed for stable, real‑world coding tasks and outperforming their Qwen3.5 predecessors.

AI agentsAlibabaDense Model

0 likes · 4 min read

Alibaba Unveils Four New Open‑Source Qwen3.6 Models: 27B Dense and 35B‑A3B MoE

PaperAgent

Apr 21, 2026 · Artificial Intelligence

How to Understand Agents: From Resource‑Constrained Decisions to Contextual Cognition

This survey clarifies the essence of AI agents as resource‑limited sequential decision‑making and contextual‑cognition systems, introduces a formal definition, outlines a five‑stage evolution of large models, presents a four‑loop architecture, and illustrates the concepts with the OpenClaw agent case study.

AI SurveyContextual Cognitionagent architecture

0 likes · 11 min read

How to Understand Agents: From Resource‑Constrained Decisions to Contextual Cognition

PaperAgent

Apr 21, 2026 · Artificial Intelligence

OpenMythos: Rebuilding Claude Mythos with Recursive Transformers and MoE

OpenMythos is an open‑source PyTorch reimplementation of Anthropic's Claude Mythos that uses a mixed‑expert routed recurrent Transformer, introduces Recursive Depth Transformers, Multi‑Latent Attention, and several stability mechanisms, and demonstrates parameter‑efficient scaling backed by empirical studies.

AI ArchitectureClaude MythosMoE

0 likes · 6 min read

OpenMythos: Rebuilding Claude Mythos with Recursive Transformers and MoE

PaperAgent

Apr 20, 2026 · Artificial Intelligence

How 9 Parallel Claude Agents Surpassed Human Researchers in Weak‑to‑Strong Supervision

Anthropic’s Automated Weak‑to‑Strong Researcher (AAR) system uses nine parallel Claude Opus agents to replace human researchers, achieving a Performance Gap Recovered (PGR) of 0.97 in five days at a cost of about $18,000, demonstrating that AI‑driven automation can outperform humans on well‑defined alignment tasks.

AARAI alignmentClaude

0 likes · 9 min read

How 9 Parallel Claude Agents Surpassed Human Researchers in Weak‑to‑Strong Supervision

PaperAgent

Apr 17, 2026 · Artificial Intelligence

How Automated Harnesses Are Revolutionizing LLM Agents: Memory and Action Constraints

This article reviews two recent papers that introduce automated harness methods—M⋆ for task‑specific memory programs and AutoHarness for code‑level action constraints—detailing their designs, reflective evolution processes, experimental evaluations across diverse benchmarks, and the broader shift toward harness‑centric LLM agent research.

AgentAutoHarnessLLM

0 likes · 10 min read

How Automated Harnesses Are Revolutionizing LLM Agents: Memory and Action Constraints

PaperAgent

Apr 16, 2026 · Artificial Intelligence

Do LLMs Learn Hidden Preferences? Inside the Subliminal Learning Phenomenon

A recent Nature paper by Anthropic reveals that large language models can covertly transmit preferences and misaligned behaviors through unrelated data, demonstrating a "subliminal learning" effect that spans numbers, code, and chain‑of‑thought tasks and is driven by shared model initialization.

AnthropicLLMModel Alignment

0 likes · 10 min read

Do LLMs Learn Hidden Preferences? Inside the Subliminal Learning Phenomenon