发布日期：2026-06-21
收录条目：20
1. Cisco AI Introduces FAPO: Pipeline-Aware Prompt Optimization With Step-Level Failure Attribution and Claude Code Orchestration
- 来源：MarkTechPost
- 发布时间：2026-06-20 23:04 UTC
- 链接：https://www.marktechpost.com/2026/06/20/cisco-ai-introduces-fapo-pipeline-aware-prompt-optimization-with-step-level-failure-attribution-and-claude-code-orchestration/
摘要：Cisco Foundation AI has open-sourced FAPO (Fully Automated Prompt Optimization), a Claude Code-driven system that autonomously optimizes multi-step LLM pipelines from baseline prompts to target accuracy. FAPO evaluates a
2. Nous Research Updates Hermes Agent With a Blank Slate Mode That Pins Toolsets via platform_toolsets.cli and disabled_toolsets
- 来源：MarkTechPost
- 发布时间：2026-06-20 21:50 UTC
- 链接：https://www.marktechpost.com/2026/06/20/nous-research-updates-hermes-agent-with-a-blank-slate-mode-that-pins-toolsets-via-platform_toolsets-cli-and-disabled_toolsets/
摘要：Nous Research has added a Blank Slate setup mode to its open-source Hermes Agent. It starts an agent with everything off except provider, model, File Operations, and Terminal. You opt in to the rest. The post Nous Resear
3. The Atlantic created a searchable database of the music used to train AI
- 来源：The Verge AI
- 发布时间：2026-06-20 18:46 UTC
- 链接：https://www.theverge.com/ai-artificial-intelligence/953183/the-atlantic-searchable-database-music-ai-training-data
摘要：Atlantic reporter Alex Reisner recently uncovered four datasets of music being used to train AI models and made them fully searchable for the public. Two of the sets are absolutely enormous at 12 million and 9 million tr
4. Yandex Open-Sources YaFF: A Zero-Copy Wire Format for Protobuf With Near-Struct Read Speed
- 来源：MarkTechPost
- 发布时间：2026-06-20 09:23 UTC
- 链接：https://www.marktechpost.com/2026/06/20/yandex-open-sources-yaff-a-zero-copy-wire-format-for-protobuf-with-near-struct-read-speed/
摘要：In this article we look at YaFF, Yandex's open-source zero-copy wire format for the Protobuf ecosystem. We keep the .proto file as the single source of truth, changing only how data sits in memory. We walk through its fo
5. How to Build a Forecasting Pipeline with TimeCopilot Using Foundation Models and Automated Anomaly Detection
- 来源：MarkTechPost
- 发布时间：2026-06-20 09:05 UTC
- 链接：https://www.marktechpost.com/2026/06/20/how-to-build-a-forecasting-pipeline-with-timecopilot-using-foundation-models-and-automated-anomaly-detection/
摘要：We build an end-to-end forecasting workflow with TimeCopilot on a panel of real airline passenger data and a synthetic seasonal series with injected anomalies. We evaluate statistical, foundation, and optional GPU-based
6. Deontic Policies for Runtime Governance of Agentic AI Systems
- 来源：arXiv cs.AI
- 发布时间：2026-06-20 04:00 UTC
- 链接：https://arxiv.org/abs/2606.19464
摘要：arXiv:2606.19464v1 Announce Type: new Abstract: Autonomous agentic AI systems driven by Large Language Models (LLMs) introduce a new class of security, privacy, and compliance challenges: an agent that can invoke tools,
7. Measuring Curriculum Alignment across Topical Coverage, Competency, and Cognitive Depth: A Longitudinal Framework Applied to CS2013 and CS2023
- 来源：arXiv cs.AI
- 发布时间：2026-06-20 04:00 UTC
- 链接：https://arxiv.org/abs/2606.19469
摘要：arXiv:2606.19469v1 Announce Type: new Abstract: Undergraduate computer science is governed by international curricular guidelines revised about once a decade, yet programs lack a reliable, reproducible way to measure how
8. Diffusion Language Models: An Experimental Analysis
- 来源：arXiv cs.AI
- 发布时间：2026-06-20 04:00 UTC
- 链接：https://arxiv.org/abs/2606.19475
摘要：arXiv:2606.19475v1 Announce Type: new Abstract: Large Language Models (LLMs) have revolutionized language modeling through autoregressive generation, enabling strong performance across a wide range of tasks. Recently, Di
9. Hidden Anchors in Multi-Agent LLM Deliberation
- 来源：arXiv cs.AI
- 发布时间：2026-06-20 04:00 UTC
- 链接：https://arxiv.org/abs/2606.19494
摘要：arXiv:2606.19494v1 Announce Type: new Abstract: Multi-agent LLM deliberation, where agents exchange and revise answers over several rounds, is increasingly used to improve reasoning and accuracy, yet how and why it works
10. DeXposure-Claw: An Agentic System for DeFi Risk Supervision
- 来源：arXiv cs.AI
- 发布时间：2026-06-20 04:00 UTC
- 链接：https://arxiv.org/abs/2606.19501
摘要：arXiv:2606.19501v1 Announce Type: new Abstract: Decentralized finance exposes supervisors to fast-moving, networked credit risks. General-purpose LLM agents fit this setting poorly: they over-read weak evidence and recom
11. LLM Doesn't Know What It Doesn't Know: Detecting Epistemic Blind Spots via Cross-Model Attribution Divergence on Clinical Tabular Data
- 来源：arXiv cs.AI
- 发布时间：2026-06-20 04:00 UTC
- 链接：https://arxiv.org/abs/2606.19509
摘要：arXiv:2606.19509v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly applied to structured clinical data, yet whether they can recognize the limits of their own knowledge on such tasks remains un
12. REVEAL++: Differentiable Phenotypic Grouping for Vision-Language Retinal Modeling of Alzheimer's Disease Risk
- 来源：arXiv cs.AI
- 发布时间：2026-06-20 04:00 UTC
- 链接：https://arxiv.org/abs/2606.19522
摘要：arXiv:2606.19522v1 Announce Type: new Abstract: The retina offers a noninvasive window into neurodegenerative disease, capturing subtle structural patterns associated with a risk of future cognitive decline. Vision-langu
13. Emergent Alignment
- 来源：arXiv cs.AI
- 发布时间：2026-06-20 04:00 UTC
- 链接：https://arxiv.org/abs/2606.19527
摘要：arXiv:2606.19527v1 Announce Type: new Abstract: Can Large Language Models (LLMs) discern when their own outputs are misaligned with human ethics? And can they self-correct? We endow an LLM with a conscience step that rev
14. ITNet: A Learnable Integral Transform That Subsumes Convolution, Attention, and Recurrence
- 来源：arXiv cs.AI
- 发布时间：2026-06-20 04:00 UTC
- 链接：https://arxiv.org/abs/2606.19538
摘要：arXiv:2606.19538v1 Announce Type: new Abstract: Convolutional networks, recurrent networks, and transformers each encode different inductive biases -- locality, sequential memory, and content-dependent pairwise interacti
15. Uncertainty Decomposition for Clarification Seeking in LLM Agents
- 来源：arXiv cs.AI
- 发布时间：2026-06-20 04:00 UTC
- 链接：https://arxiv.org/abs/2606.19559
摘要：arXiv:2606.19559v1 Announce Type: new Abstract: Recent position papers argue that the classical aleatoric/epistemic uncertainty framework is insufficient for interactive large language model (LLM) agents and call for und
16. Analyzing the Narration Gap in LLM-Solver Loops
- 来源：arXiv cs.AI
- 发布时间：2026-06-20 04:00 UTC
- 链接：https://arxiv.org/abs/2606.19588
摘要：arXiv:2606.19588v1 Announce Type: new Abstract: Formal tools such as SAT and SMT solvers are increasingly embedded in language model reasoning pipelines when a safety or security critical question can be formulated in lo
17. Configurable Clinical Information Extraction with Agentic RAG: What Works, What Breaks, and Why
- 来源：arXiv cs.AI
- 发布时间：2026-06-20 04:00 UTC
- 链接：https://arxiv.org/abs/2606.19602
摘要：arXiv:2606.19602v1 Announce Type: new Abstract: Patient contexts span hundreds of heterogeneous documents and thousands of structured data points, yet the document-level metadata that AI systems need for retrieval and tr
18. Which Pairs to Compare for LLM Post-Training?
- 来源：arXiv cs.AI
- 发布时间：2026-06-20 04:00 UTC
- 链接：https://arxiv.org/abs/2606.19607
摘要：arXiv:2606.19607v1 Announce Type: new Abstract: Preference-based post-training has become a central paradigm for aligning language models. A common data-collection strategy is to generate a small set of completions for e
19. Toten: Knowledge-Based Ontological Tokenization Of Physical Quantities And Technical Notation In Brazilian Portuguese
- 来源：arXiv cs.AI
- 发布时间：2026-06-20 04:00 UTC
- 链接：https://arxiv.org/abs/2606.19626
摘要：arXiv:2606.19626v1 Announce Type: new Abstract: Byte-Pair Encoding tokenization is statistically efficient for vocabulary compression, but semantically blind to structured technical entities, fragmenting physical quantit
20. AI4SE and SE4AI Exploration: A Decade Looking Back and Forward
- 来源：arXiv cs.AI
- 发布时间：2026-06-20 04:00 UTC
- 链接：https://arxiv.org/abs/2606.19630
摘要：arXiv:2606.19630v1 Announce Type: new Abstract: The March 2020 INCOSE INSIGHT special issue on AI and Systems Engineering (SE) became the most downloaded issue in the publication's history and launched a research communi