# 每日论文简报

- 生成时间：2026-04-11 23:09:08 (Asia/Shanghai)
- 检索窗口：最近 24 小时
- 命中概览：LLM=0, Vision=0, PubMed AI=4, OpenAlex AI=5
- 排序策略：hybrid (relevance first, published_at tie-break)

## 今日重点

- 主题「Agent」：命中 5 篇，覆盖 OpenAlex AI，代表论文包括 《Coalition Drift: When Agents Drift Together Why multi-agent systems don't just drift individually — they drift as a group, and why that matters more than any single-agent failure mode.》、《Coalition Formation Events: How Multi-Agent Systems Create Temporary Actors》。
- 主题「Clinical」：命中 4 篇，覆盖 PubMed AI，代表论文包括 《Evaluating the clinical decision-making performance of large language models in clinically oriented thoracic anatomy scenarios: a comparative evaluation study.》、《Exploratory study of large language models in surgical decision-making for lumbar disc herniation: a multicenter analysis based on multisource clinical information.》。
- 主题「Language Model」：命中 4 篇，覆盖 PubMed AI，代表论文包括 《Evaluating the clinical decision-making performance of large language models in clinically oriented thoracic anatomy scenarios: a comparative evaluation study.》、《Exploratory study of large language models in surgical decision-making for lumbar disc herniation: a multicenter analysis based on multisource clinical information.》。

## 主题聚焦

### Agent

- 命中篇数：5
- 覆盖分组：OpenAlex AI
- 代表论文：《Coalition Drift: When Agents Drift Together Why multi-agent systems don't just drift individually — they drift as a group, and why that matters more than any single-agent failure mode.》、《Coalition Formation Events: How Multi-Agent Systems Create Temporary Actors》、《U-P Duality in Multi-Agent Systems: A Seven-Space Algorithm for Complex Nonlinear AI (Corrected Version)》
- 主题速读：
  - 《Coalition Drift: When Agents Drift Together Why multi-agent systems don't just drift individually — they drift as a group, and why that matters more than any single-agent failure mode.》〔方法〕：Most AI governance frameworks still treat drift as an individual phenomenon: one agent misinterprets, one agent hallucinates, one agent misroutes, one agent co…
  - 《Coalition Formation Events: How Multi-Agent Systems Create Temporary Actors》〔应用 / 方法〕：In multi-agent ecosystems, the meaningful unit of behavior is not the individual agent. It is the coalition — the temporary, emergent structure formed when mul…

### Clinical

- 命中篇数：4
- 覆盖分组：PubMed AI
- 代表论文：《Evaluating the clinical decision-making performance of large language models in clinically oriented thoracic anatomy scenarios: a comparative evaluation study.》、《Exploratory study of large language models in surgical decision-making for lumbar disc herniation: a multicenter analysis based on multisource clinical information.》、《A hybrid large language model framework for structured data entry from code-switched persian clinical speech.》
- 主题速读：
  - 《Evaluating the clinical decision-making performance of large language models in clinically oriented thoracic anatomy scenarios: a comparative evaluation study.》〔评测 / 方法〕：
  - 《Exploratory study of large language models in surgical decision-making for lumbar disc herniation: a multicenter analysis based on multisource clinical information.》〔方法〕：

### Language Model

- 命中篇数：4
- 覆盖分组：PubMed AI
- 代表论文：《Evaluating the clinical decision-making performance of large language models in clinically oriented thoracic anatomy scenarios: a comparative evaluation study.》、《Exploratory study of large language models in surgical decision-making for lumbar disc herniation: a multicenter analysis based on multisource clinical information.》、《A hybrid large language model framework for structured data entry from code-switched persian clinical speech.》
- 主题速读：
  - 《Evaluating the clinical decision-making performance of large language models in clinically oriented thoracic anatomy scenarios: a comparative evaluation study.》〔评测 / 方法〕：
  - 《Exploratory study of large language models in surgical decision-making for lumbar disc herniation: a multicenter analysis based on multisource clinical information.》〔方法〕：

## LLM 观察

今日没有新的命中文献。

## Vision 观察

今日没有新的命中文献。

## PubMed AI 观察

### 本组速览

- 《Factors influencing large language model adoption among dental students: a cross-sectional study.》〔应用 / 方法〕：This research evaluates the factors influencing the behavioural intention (BI) to adopt large language models (LLMs) among dental students in education, clinic…

### 论文速览

1. [Evaluating the clinical decision-making performance of large language models in clinically oriented thoracic anatomy scenarios: a comparative evaluation study.](https://pubmed.ncbi.nlm.nih.gov/41963950/)
   - Entered：2026-04-11 07:53
   - 作者：Zeynep Nisa Karakoyun，Mustafa Deniz Yörük，Mehmed Emre Özdemir，Mehmet İlkay Koşar
   - 来源：pubmed
   - 相关性分数：104
   - 命中原因：title matched "language model"; title matched "clinical"; has DOI; has complete metadata
   - 分类：Journal Article
   - 标签：评测 / 方法
   - 主题词：Clinical / Language Model
   - 摘要：

2. [Exploratory study of large language models in surgical decision-making for lumbar disc herniation: a multicenter analysis based on multisource clinical information.](https://pubmed.ncbi.nlm.nih.gov/41963879/)
   - Entered：2026-04-11 07:49
   - 作者：Zebing Ma，Liang Kong，Yang Shu，Lili Ni，Yongxin Ma，Rui Chen 等
   - 来源：pubmed
   - 相关性分数：104
   - 命中原因：title matched "language model"; title matched "clinical"; has DOI; has complete metadata
   - 分类：Journal Article
   - 标签：方法
   - 主题词：Clinical / Language Model
   - 摘要：

3. [A hybrid large language model framework for structured data entry from code-switched persian clinical speech.](https://pubmed.ncbi.nlm.nih.gov/41963402/)
   - Entered：2026-04-11 07:20
   - 作者：Mohammad Tanhaei
   - 来源：pubmed
   - 相关性分数：104
   - 命中原因：title matched "language model"; title matched "clinical"; has DOI; has complete metadata
   - 分类：Journal Article
   - 标签：方法
   - 主题词：Clinical / Language Model
   - 摘要：

4. [Factors influencing large language model adoption among dental students: a cross-sectional study.](https://pubmed.ncbi.nlm.nih.gov/41963506/)
   - Entered：2026-04-11 07:24
   - 作者：Angel M Joseph，Rahaf E Almutairi，Wejdan A Alrashidi，Reem M Almutairi，Yahya Theeban，Sami Aldhuwayhi 等
   - 来源：pubmed
   - 相关性分数：88
   - 命中原因：title matched "language model"; summary matched "clinical"; has DOI; has rich summary
   - 分类：Journal Article
   - 标签：应用 / 方法
   - 主题词：Clinical / Language Model
   - 摘要：This research evaluates the factors influencing the behavioural intention (BI) to adopt large language models (LLMs) among dental students in education, clinical decision support (CDS), and research, using the original unified theory of acceptance and use of technology (UTAUT) model, representing the first application of this model in this specific context. LLM adoption among Saudi dental students is unstructured and unregulated, making empirical evidence on adoption factors an educational and governance priority. A cross-sectional, web-based survey using convenience sampling was conducted among 700 students from six universities in central Saudi Arabia (60.71% aged 18-22 years; 58.71% female; 74% undergraduate). The UTAUT constructs: performance expectancy (PE), effort expectancy (EE), social influence (SI), and facilitating conditions (FC), were assessed using a five-point Likert scale. Measurement equivalence across English and Arabic language questionnaire versions were confirmed. All UTAUT constructs showed borderline acceptable to excellent internal consistency (α = 0.678-0.907). Structural equation modelling (SEM) with bootstrapped 95% confidence intervals (CI) was used for statistical analysis. Hierarchical multiple regression was used for post-hoc sensitivity analysis to examine the moderating effects of demographic variables. The model accounted for 44.1% of the variance in BI (SEM-derived R 2 = 0.441, p < 0.001). EE was the strongest significant factor influencing BI (β = 0.351, p = 0.017, 95% CI [0.056, 0.518]), followed by FC (β = 0.136, p = 0.024, 95% CI [0.019, 0.183]). The PE (β = 0.158, p = 0.162) was non-significant, due to its high latent correlation with EE, and SI (β = 0.090, p = 0.131) was also non-significant, consistent with absent institutional mandates. The UTAUT-BI relationships were not significantly moderated by age, gender, or academic level. The results indicate strong expressed BI to adopt LLMs among dental students; however, BI is a theoretical proxy and should not be equated with actual adoption behaviour, clinical competence, or ethical and effective LLM application. These findings offer preliminary evidence which may inform curriculum design and institutional governance for responsible LLM integration in dentistry. The tentatively proposed LLM competency framework represents theoretical concepts which require empirical validation before implementation.

## OpenAlex AI 观察

### 本组速览

- 《Coalition Drift: When Agents Drift Together Why multi-agent systems don't just drift individually — they drift as a group, and why that matters more than any single-agent failure mode.》〔方法〕：Most AI governance frameworks still treat drift as an individual phenomenon: one agent misinterprets, one agent hallucinates, one agent misroutes, one agent co…
- 《Coalition Drift: When Agents Drift Together Why multi-agent systems don't just drift individually — they drift as a group, and why that matters more than any single-agent failure mode.》〔方法〕：Most AI governance frameworks still treat drift as an individual phenomenon: one agent misinterprets, one agent hallucinates, one agent misroutes, one agent co…
- 《Coalition Formation Events: How Multi-Agent Systems Create Temporary Actors》〔应用 / 方法〕：In multi-agent ecosystems, the meaningful unit of behavior is not the individual agent. It is the coalition — the temporary, emergent structure formed when mul…

### 论文速览

1. [Coalition Drift: When Agents Drift Together Why multi-agent systems don't just drift individually — they drift as a group, and why that matters more than any single-agent failure mode.](https://doi.org/10.5281/zenodo.19502960)
   - Published：2026-04-11 08:00
   - 作者：Narnaiezzsshaa Truong
   - 来源：openalex
   - 相关性分数：70
   - 命中原因：title matched "agent"; has DOI; has rich summary; has complete metadata
   - 分类：Article, Physics and Astronomy, Statistical and Nonlinear Physics, Opinion Dynamics and Social Influence, Game Theory and Applications
   - 标签：方法
   - 主题词：Agent
   - 摘要：Most AI governance frameworks still treat drift as an individual phenomenon: one agent misinterprets, one agent hallucinates, one agent misroutes, one agent compensates, one agent bypasses. This framing works only when agents operate in isolation. Modern systems are multi-agent ecosystems: agents coordinate, delegate, synchronize, and share context. In these environments, drift is no longer an individual deviation. It becomes a coalition-level phenomenon. Coalition drift is what happens when agents — each behaving correctly according to their local rules — collectively drift in the same direction, producing a coordinated but incorrect system-level outcome. This paper explains what coalition drift is, why it emerges, how it propagates, and why it is the most dangerous form of drift in multi-agent systems. Individual drift is a nuisance. Coalition drift is an existential governance risk.

2. [Coalition Drift: When Agents Drift Together Why multi-agent systems don't just drift individually — they drift as a group, and why that matters more than any single-agent failure mode.](https://doi.org/10.5281/zenodo.19502961)
   - Published：2026-04-11 08:00
   - 作者：Narnaiezzsshaa Truong
   - 来源：openalex
   - 相关性分数：70
   - 命中原因：title matched "agent"; has DOI; has rich summary; has complete metadata
   - 分类：Article, Physics and Astronomy, Statistical and Nonlinear Physics, Opinion Dynamics and Social Influence, Game Theory and Applications
   - 标签：方法
   - 主题词：Agent
   - 摘要：Most AI governance frameworks still treat drift as an individual phenomenon: one agent misinterprets, one agent hallucinates, one agent misroutes, one agent compensates, one agent bypasses. This framing works only when agents operate in isolation. Modern systems are multi-agent ecosystems: agents coordinate, delegate, synchronize, and share context. In these environments, drift is no longer an individual deviation. It becomes a coalition-level phenomenon. Coalition drift is what happens when agents — each behaving correctly according to their local rules — collectively drift in the same direction, producing a coordinated but incorrect system-level outcome. This paper explains what coalition drift is, why it emerges, how it propagates, and why it is the most dangerous form of drift in multi-agent systems. Individual drift is a nuisance. Coalition drift is an existential governance risk.

3. [Coalition Formation Events: How Multi-Agent Systems Create Temporary Actors](https://doi.org/10.5281/zenodo.19503043)
   - Published：2026-04-11 08:00
   - 作者：Narnaiezzsshaa Truong
   - 来源：openalex
   - 相关性分数：70
   - 命中原因：title matched "agent"; has DOI; has rich summary; has complete metadata
   - 分类：Article, Computer Science, Artificial Intelligence, Multi-Agent Systems and Negotiation, Evolutionary Game Theory and Cooperation
   - 标签：应用 / 方法
   - 主题词：Agent
   - 摘要：In multi-agent ecosystems, the meaningful unit of behavior is not the individual agent. It is the coalition — the temporary, emergent structure formed when multiple agents synchronize around shared context, shared assumptions, and shared workflows. Coalitions are not explicitly created. They form themselves. Coalition formation is not a design choice. It is a physics event. This paper explains how coalitions form, why they behave like temporary actors, and what governance must observe to understand them. Coalition formation is the beginning of drift. Coalition dissolution is the end of drift. Governance must capture both — and everything in between. The formation event is where governance intervention has the highest leverage.

4. [Coalition Formation Events: How Multi-Agent Systems Create Temporary Actors](https://doi.org/10.5281/zenodo.19503044)
   - Published：2026-04-11 08:00
   - 作者：Narnaiezzsshaa Truong
   - 来源：openalex
   - 相关性分数：70
   - 命中原因：title matched "agent"; has DOI; has rich summary; has complete metadata
   - 分类：Article, Computer Science, Artificial Intelligence, Multi-Agent Systems and Negotiation, Evolutionary Game Theory and Cooperation
   - 标签：应用 / 方法
   - 主题词：Agent
   - 摘要：In multi-agent ecosystems, the meaningful unit of behavior is not the individual agent. It is the coalition — the temporary, emergent structure formed when multiple agents synchronize around shared context, shared assumptions, and shared workflows. Coalitions are not explicitly created. They form themselves. Coalition formation is not a design choice. It is a physics event. This paper explains how coalitions form, why they behave like temporary actors, and what governance must observe to understand them. Coalition formation is the beginning of drift. Coalition dissolution is the end of drift. Governance must capture both — and everything in between. The formation event is where governance intervention has the highest leverage.

5. [U-P Duality in Multi-Agent Systems: A Seven-Space Algorithm for Complex Nonlinear AI (Corrected Version)](https://doi.org/10.5281/zenodo.19163012)
   - Published：2026-04-11 08:00
   - 作者：Xie Meng
   - 来源：openalex
   - 相关性分数：70
   - 命中原因：title matched "agent"; has DOI; has rich summary; has complete metadata
   - 分类：Preprint, Computer Science, Computational Theory and Mathematics, Adaptive Dynamic Programming Control, Distributed Control Multi-Agent Systems
   - 标签：应用 / 方法
   - 主题词：Agent
   - 摘要：Multi-agent systems are a core research area in distributed artificial intelligence, where algorithm design faces challenges due to complex nonlinear environments. This paper presents a seven-space algorithm based on U-P duality, including U-space for stabilization, P-space for asynchronous processing, V-space for exploration, Z-space for adaptive regulation, Q-space for anomaly detection, R-space for redundancy backup, and S-space for scene understanding. Each space is derived from a common fixed-point relation, with behavior regulated by a parameter s. Numerical experiments in autonomous driving perception, distributed training, and reinforcement learning show that near the critical line, P-space achieves accurate multi-sensor fusion, P+R combination achieves robust fault tolerance, and V-space provides balanced exploration-exploitation. Derivative relations provide closed-form expressions for gradient-based optimization. The seven spaces constitute a set of algorithms for nonlinear AI domains, with mathematical guarantees from fixed-point theory. Users can select parameters according to specific scenarios to obtain adapted behaviors. Correction Note: This corrected version updates the U-P duality formula from W_{-1}(-A) to W_0(-A), and the Z-Q duality formula accordingly. The P-space fixed point is dynamically accessible only through the principal branch W_0. All numerical results and algorithmic procedures remain unchanged.