# Research Radar · 2026-05-23 (Sat)

_Auto-generated daily arXiv monitoring dashboard_

## 🔬 Fine-Grained CV (5 papers)

### 1. Training-Free Fine-Grained Semantic Segmentations in Low Data Regimes: A FungiTastic Baseline

- **arXiv:** `2605.22492` | [Abstract](https://arxiv.org/abs/2605.22492v1) | [PDF](https://arxiv.org/pdf/2605.22492v1)
- **Authors:** Sebastian Cavada, Francesco Pelosin, Lapo Faggi
- **Published:** 2026-05-21

Fine-grained semantic segmentation requires both precise localization and discrimination between visually similar classes. In FungiTastic, this problem is further complicated by a long-tailed distribution and strong variation in image acquisition conditions. We propose a training-free two-stage fram

---

### 2. Making the Discrete Continuous: Synthetic RAW Augmentations for Fine-Grained Evaluation of Person Detection Performance in Low Light

- **arXiv:** `2605.22455` | [Abstract](https://arxiv.org/abs/2605.22455v1) | [PDF](https://arxiv.org/pdf/2605.22455v1)
- **Authors:** Valeria Pais, Malena Mendilaharzu, Daniele Faccio, Luis Oala, Christoph Clausen et al.
- **Published:** 2026-05-21

Real-world deployment of AI vision models is both fueled and limited by the data available for training and testing. Real datasets are sparse and uneven: long-tailed or unbalanced distributions hinder generalization, and the low number of samples in low density regions makes it hard to run evaluatio

---

### 3. DecQ: Detail-Condensing Queries for Enhanced Reconstruction and Generation in Representation Autoencoders

- **arXiv:** `2605.22777` | [Abstract](https://arxiv.org/abs/2605.22777v1) | [PDF](https://arxiv.org/pdf/2605.22777v1)
- **Authors:** Tianhang Wang, Yitong Chen, Wei Song, Zuxuan Wu, Min Li et al.
- **Published:** 2026-05-21

Representation Autoencoders (RAEs) leverage frozen vision foundation models (VFMs) as tokenizer encoders, providing robust high-level representations that facilitate fast convergence and high-quality generation in latent diffusion models. However, freezing the VFM inherently constrains its spatial r

---

### 4. Improving Viewpoint-Invariance and Temporal Consistency for Action Detection

- **arXiv:** `2605.22695` | [Abstract](https://arxiv.org/abs/2605.22695v1) | [PDF](https://arxiv.org/pdf/2605.22695v1)
- **Authors:** Yannick Porto, Renato Martins, Thomas Chalumeau, Cedric Demonceaux
- **Published:** 2026-05-21

Viewpoint change invariance and action temporal consistency are critical aspects for the effective deployment of human action detection of untrimmed videos. Existing appearance-based video detection methods often struggle with limited viewpoint diversity during training, while motion-based detection

---

### 5. AtomicMotion: Learning Human Motion From Different Human Parts

- **arXiv:** `2605.22631` | [Abstract](https://arxiv.org/abs/2605.22631v1) | [PDF](https://arxiv.org/pdf/2605.22631v1)
- **Authors:** Runzhen Liu, Chuhua Xian, Fa-Ting Hong
- **Published:** 2026-05-21

Accurately reconstructing full-body poses from sparse head and hand trajectories is a foundational challenge for immersive AR/VR telepresence. Current methods often struggle with error accumulation and unnatural joint coordination, primarily because they treat the human body as a monolithic entity, 

---

## 📊 CTR / Ranking / IR (5 papers)

### 1. Diversed Model Discovery via Structured Table Discovery

- **arXiv:** `2605.22766` | [Abstract](https://arxiv.org/abs/2605.22766v1) | [PDF](https://arxiv.org/pdf/2605.22766v1)
- **Authors:** Zhengyuan Dong, Renée J. Miller
- **Published:** 2026-05-21

Model cards describe model behavior through a mixture of textual descriptions and structured artifacts, including performance, configuration, and dataset tables. Existing model search systems rely predominantly on semantic similarity over text, which can produce homogeneous result sets and limit exp

---

### 2. Search-E1: Self-Distillation Drives Self-Evolution in Search-Augmented Reasoning

- **arXiv:** `2605.22511` | [Abstract](https://arxiv.org/abs/2605.22511v1) | [PDF](https://arxiv.org/pdf/2605.22511v1)
- **Authors:** Zihan Liang, Yufei Ma, Ben Chen, Zhipeng Qian, Xuxin Zhang et al.
- **Published:** 2026-05-21

Post-training has become the dominant recipe for turning a language model into a competent search-augmented reasoning agent. A line of recent work pushes its performance further by adding elaborate machinery on top of this standard pipeline. These augmentations import external supervision from stron

---

### 3. BeLink: Biomedical Entity Linking Meets Generative Re-Ranking

- **arXiv:** `2605.22501` | [Abstract](https://arxiv.org/abs/2605.22501v1) | [PDF](https://arxiv.org/pdf/2605.22501v1)
- **Authors:** Darya Shlyk, Stefano Montanelli, Lawrence Hunter
- **Published:** 2026-05-21

Despite recent progress, Biomedical Entity Linking (BEL) with large language models (LLMs) remains computationally inefficient and challenging to deploy in practical settings. In this work, we demonstrate that instruction-tuning of open-source generative models can offer an effective solution when a

---

### 4. Behavior-Guided Candidate Calibration for Multimodal Recommendation

- **arXiv:** `2605.22073` | [Abstract](https://arxiv.org/abs/2605.22073v1) | [PDF](https://arxiv.org/pdf/2605.22073v1)
- **Authors:** Zesheng Li, Chengchang Pan, Honggang Qi
- **Published:** 2026-05-21

Multimodal recommendation benefits from content signals, but the gain depends on how those signals interact with the ranking pipeline. We find that moderate cross-view agreement helps, while stronger agreement suppresses recommendation-specific variation. Spectral analysis shows a clear split: low-f

---

### 5. Generative Conversational Recommender System

- **arXiv:** `2605.21987` | [Abstract](https://arxiv.org/abs/2605.21987v1) | [PDF](https://arxiv.org/pdf/2605.21987v1)
- **Authors:** Sixiao Zhang, Mingrui Liu, Cheng Long
- **Published:** 2026-05-21

Conversational recommender systems aim to provide personalized recommendations via natural language interactions. However, existing approaches either decouple recommendation from dialog generation or rely on retrieval-based pipelines, limiting the integration between recommendation and response gene

---

## 🤖 Agent / Auto Research (5 papers)

### 1. MOSS: Self-Evolution through Source-Level Rewriting in Autonomous Agent Systems

- **arXiv:** `2605.22794` | [Abstract](https://arxiv.org/abs/2605.22794v1) | [PDF](https://arxiv.org/pdf/2605.22794v1)
- **Authors:** Qianshu Cai, Yonggang Zhang, Xianzhang Jia, Wei Xue, Jun Song et al.
- **Published:** 2026-05-21

Autonomous agentic systems are largely static after deployment: they do not learn from user interactions, and recurring failures persist until the next human-driven update ships a fix. Self-evolving agents have emerged in response, but all confine evolution to text-mutable artifacts -- skill files, 

---

### 2. LCGuard: Latent Communication Guard for Safe KV Sharing in Multi-Agent Systems

- **arXiv:** `2605.22786` | [Abstract](https://arxiv.org/abs/2605.22786v1) | [PDF](https://arxiv.org/pdf/2605.22786v1)
- **Authors:** Sadia Asif, Mohammad Mohammadi Amiri, Momin Abbas, Prasanna Sattigeri, Karthikeyan Natesan Ramamurthy
- **Published:** 2026-05-21

Large language model (LLM)-based multi-agent systems increasingly rely on intermediate communication to coordinate complex tasks. While most existing systems communicate through natural language, recent work shows that latent communication, particularly through transformer key-value (KV) caches, can

---

### 3. DeltaBox: Scaling Stateful AI Agents with Millisecond-Level Sandbox Checkpoint/Rollback

- **arXiv:** `2605.22781` | [Abstract](https://arxiv.org/abs/2605.22781v1) | [PDF](https://arxiv.org/pdf/2605.22781v1)
- **Authors:** Yunpeng Dong, Jingkai He, Yuze Hou, Dong Du, Zhonghu Xu et al.
- **Published:** 2026-05-21

LLM-powered AI agents require high-frequency state exploration (e.g., test-time tree search and reinforcement learning), relying on rapid checkpoint and rollback (C/R) of the complete sandbox state, including files and process state (e.g., memory, contexts, etc.). Existing mechanisms duplicate the e

---

### 4. Deep Reinforcement Learning for Flexible Job Shop Scheduling with Random Job Arrivals

- **arXiv:** `2605.22773` | [Abstract](https://arxiv.org/abs/2605.22773v1) | [PDF](https://arxiv.org/pdf/2605.22773v1)
- **Authors:** Yu Tang, Muhammad Zakwan, Efe Balta, John Lygeros, Alisa Rupenyan
- **Published:** 2026-05-21

The Flexible Job Shop Scheduling Problem (FJSP) is the optimal allocation of a set of jobs to machines. Two primary challenges persist in FJSP: the unpredictable arrival of future jobs and the combinatorial complexity of the problem, rendering it intractable for conventional mixed-integer linear pro

---

### 5. Advancing Mathematics Research with AI-Driven Formal Proof Search

- **arXiv:** `2605.22763` | [Abstract](https://arxiv.org/abs/2605.22763v1) | [PDF](https://arxiv.org/pdf/2605.22763v1)
- **Authors:** George Tsoukalas, Anton Kovsharov, Sergey Shirobokov, Anja Surina, Moritz Firsching et al.
- **Published:** 2026-05-21

Large language models (LLMs) increasingly excel at mathematical reasoning, but their unreliability limits their utility in mathematics research. A mitigation is using LLMs to generate formal proofs in languages like Lean. We perform the first large-scale evaluation of this method's ability to solve 

---

_Report generated at 2026-05-23 16:22:06 CST_