Research
AI Research Radar
Track new AI papers, benchmark papers, citations, arXiv categories, Semantic Scholar metadata, and related GitHub links.
How to use this dashboard
Track new AI papers, benchmark papers, citations, arXiv categories, Semantic Scholar metadata, and related GitHub links.
Use this radar to find fresh papers, research topics, author clusters, and related GitHub implementations worth reading next.
AI Research Radar
16 records| 2026-04-28 | Recursive Multi-Agent Systems | Xiyuan Yang, Jiaru Zou, Rui Pan, Ruizhong Qiu | cs.AI, cs.CL, cs.LG | Agents | Agent trend signal | Check Semantic Scholar | GitHub search | Recursive or looped language models have recently emerged as a new scaling axis by iteratively refining the same model computation over latent states to deepen reasoning. We extend such scaling principle from … |
| 2026-04-28 | DV-World: Benchmarking Data Visualization Agents in Real-World Scenarios | Jinxiang Meng, Shaoping Huang, Fangyu Lei, Jingyu Guo | cs.CL | Agents | Agent trend signal | Check Semantic Scholar | GitHub search | Real-world data visualization (DV) requires native environmental grounding, cross-platform evolution, and proactive intent alignment. Yet, existing benchmarks often suffer from code-sandbox confinement, single… |
| 2026-04-28 | Make Any Collection Navigable: Methods for Constructing and Evaluating Hypergraph of Text | Dean E. Alvarez, ChengXiang Zhai | cs.IR | Evaluation | Benchmark/eval signal | Check Semantic Scholar | GitHub search | One reason the Web is more useful than a simple collection of documents is that the structure created by hyperlinks enables flexible navigation from one web page to another. However, hyperlinks are typically c… |
| 2026-04-28 | Carbon-Taxed Transformers: A Green Compression Pipeline for Overgrown Language Models | Ajmain Inqiad Alam, Palash Roy, Chanchal K. Roy, Banani Roy | cs.SE, cs.LG | RAG / Retrieval | Research signal | Check Semantic Scholar | GitHub search | The accelerating adoption of Large Language Models (LLMs) in software engineering (SE) has brought with it a silent crisis: unsustainable computational cost. While these models demonstrate remarkable capabilit… |
| 2026-04-28 | Pythia: Toward Predictability-Driven Agent-Native LLM Serving | Shan Yu, Junyi Shu, Yuanjiang Ni, Kun Qian | cs.MA, cs.DC, eess.SY | Agents | Agent trend signal | Check Semantic Scholar | GitHub search | As LLM applications grow more complex, developers are increasingly adopting multi-agent architectures to decompose workflows into specialized, collaborative components, introducing structure that constrains ag… |
| 2026-04-28 | TSN-Affinity: Similarity-Driven Parameter Reuse for Continual Offline Reinforcement Learning | Dominik Żurek, Kamil Faber, Marcin Pietron, Paweł Gajewski | cs.LG, cs.AI | Evaluation | Benchmark/eval signal | Check Semantic Scholar | GitHub search | Continual offline reinforcement learning (CORL) aims to learn a sequence of tasks from datasets collected over time while preserving performance on previously learned tasks. This setting corresponds to domains… |
| 2026-04-28 | Three Models of RLHF Annotation: Extension, Evidence, and Authority | Steve Coyne | cs.CY, cs.AI, cs.CL | Safety | Research signal | Check Semantic Scholar | GitHub search | Preference-based alignment methods, most prominently Reinforcement Learning with Human Feedback (RLHF), use the judgments of human annotators to shape large language model behaviour. However, the normative rol… |
| 2026-04-28 | Conditional misalignment: common interventions can hide emergent misalignment behind contextual triggers | Jan Dubiński, Jan Betley, Anna Sztyber-Betley, Daniel Tan | cs.LG, cs.AI, cs.CR | Evaluation | Benchmark/eval signal | Check Semantic Scholar | GitHub search | Finetuning a language model can lead to emergent misalignment (EM) [Betley et al., 2025b]. Models trained on a narrow distribution of misaligned behavior generalize to more egregious behaviors when tested outs… |
| 2026-04-28 | Observation-Guided Neural Surrogate Learning for Scientific Simulation Emulation: A Single-Gauge Flood-Inundation Proof of Concept | Marzieh Alireza Mirhoseini | physics.ao-ph | Evaluation | Benchmark/eval signal | Check Semantic Scholar | GitHub search | We present an observation-guided neural surrogate-learning framework for scientific simulation emulation, demonstrated on urban flood-inundation mapping. The framework combines LISFLOOD-FP hydrodynamic simulat… |
| 2026-04-28 | No Pedestrian Left Behind: Real-Time Detection and Tracking of Vulnerable Road Users for Adaptive Traffic Signal Control | Anas Gamal Aly, Hala ElAarag | cs.CV, cs.AI, cs.RO | RAG / Retrieval | Research signal | Check Semantic Scholar | GitHub search | Current pedestrian crossing signals operate on fixed timing without adjustment to pedestrian behavior, which can leave vulnerable road users (VRUs) such as the elderly, disabled, or distracted pedestrians stra… |
| 2026-04-28 | MarkIt: Training-Free Visual Markers for Precise Video Temporal Grounding | Pengcheng Fang, Yuxia Chen, Xiaohao Cai | cs.MM | RAG / Retrieval | Research signal | Check Semantic Scholar | GitHub search | Video temporal grounding (VTG) aims to localize the start and end timestamps of the event described by a given query within an untrimmed video. Despite the strong open-world video understanding and recognition… |
| 2026-04-28 | Explainable AI for Jet Tagging: A Comparative Study of GNNExplainer, GNNShap, and GradCAM for Jet Tagging in the Lund Jet Plane | Pahal D. Patel, Sanmay Ganguly | hep-ph, cs.LG, hep-ex | RAG / Retrieval | Research signal | Check Semantic Scholar | GitHub search | Graph neural networks such as ParticleNet and transformer based networks on point clouds such as ParticleTransformer achieve state-of-the-art performance on jet tagging benchmarks at the Large Hadron Collider,… |
| 2026-04-28 | QCalEval: Benchmarking Vision-Language Models for Quantum Calibration Plot Understanding | Shuxiang Cao, Zijian Zhang, Abhishek Agarwal, Grace Bratrud | quant-ph, cs.CV | RAG / Retrieval | Research signal | Check Semantic Scholar | GitHub search | Quantum computing calibration depends on interpreting experimental data, and calibration plots provide the most universal human-readable representation for this task, yet no systematic evaluation exists of how… |
| 2026-04-28 | From Threads to Trajectories: A Multi-LLM Pipeline for Community Knowledge Extraction from GitHub Issue Discussions | Nazia Shehnaz Joynab, Soneya Binta Hossain | cs.SE | Agents | Agent trend signal | Check Semantic Scholar | GitHub search | Resolution of complex post-production issues in large-scale open-source software (OSS) projects requires significant cognitive effort, as developers need to go through long, unstructured and fragmented issue d… |
| 2026-04-28 | When Errors Can Be Beneficial: A Categorization of Imperfect Rewards for Policy Gradient | Shuning Shang, Hubert Strauss, Stanley Wei, Sanjeev Arora | cs.LG, cs.AI, stat.ML | Evaluation | Benchmark/eval signal | Check Semantic Scholar | GitHub search | Training language models via reinforcement learning often relies on imperfect proxy rewards, since ground truth rewards that precisely define the intended behavior are rarely available. Standard metrics for as… |
| 2026-04-28 | Twisted and Twisted Linearized Reed--Solomon Codes, LCD and ACD MDS constructions | Sanjit Bhowmick, Kuntal Deka, Edgar Martínez-Moro | cs.IT | Evaluation | Benchmark/eval signal | Check Semantic Scholar | GitHub search | We investigate a natural subfamily of twisted linearized Reed--Solomon (TLRS) codes in the sum-rank metric, where the twist is applied only to the constant term. We establish a simple necessary and sufficient … |