Tag
Keenable SELECT is an MCP server that enables AI agents to execute SQL queries on live web data, using semantic operators for efficient search and extraction, and generating HTML reports.
Lieflat Charts is an open-source tool that has reached 2k GitHub stars, enabling AI agents to create aesthetically pleasing charts with over 60 types using HTML for rich animations and report generation.
Building an AI report generator for a client revealed that most effort went into data normalization and templating, with the model being a minor component, underscoring the primacy of data pipelines over generation.
The author built an AI agent for market research that automates report assembly but still requires human judgment for contextual decisions, highlighting the balance between AI automation and human expertise in knowledge work.
RadFusion is a framework that adds threshold controllability to radiology report generation by fusing a multi-label classifier with a VQA-based generator and an LLM rewrite step, enabling sensitivity-specificity trade-offs and ROC-based validation. Experiments on MIMIC-CXR show improved diagnostic accuracy and clinically adaptable report behavior.
A developer shares lessons from running an AI report generator in production, arguing that data quality and validation matter far more than the model's writing ability, since fluent but incorrect reports are dangerous.
Introduces MCTS-Report, a Monte Carlo Tree Search framework for generating multimodal reports from tabular data, along with the MMRBench benchmark. It outperforms strong baselines across structural completeness, numerical accuracy, chart-text alignment, and insight novelty.
This paper highlights that VLMs for chest x-ray report generation can score well on benchmarks while erasing clinically meaningful terms and introducing biased language, and proposes a framework to measure these failures.
This paper adapts a diffusion language model for interactive radiology report drafting, showing it matches autoregressive models in accuracy while offering unique infill capabilities that allow radiologists to fix report fragments and have the model fill in the text between them.
A tool that automates research and report generation by aggregating information from multiple sources, likely using AI.
This paper introduces transition-aware best-of-N sampling, a training-free method for generating longitudinal chest X-ray reports by encoding changes between prior and current examinations using set-to-set distance metrics.
This paper investigates parameter-efficient strategies for adapting large language models to 3D CT report generation, introducing RAD3D-Prefix, a lightweight diagnostic-prior conditioning framework that keeps the LLM frozen and requires minimal trainable parameters. It shows that freezing larger LLMs (~1B+) and training only lightweight projection layers provides a superior trade-off between performance, generalization, and computational efficiency.
An open-source project providing an Opencode Skill that automatically generates in-depth research reports comparable to those from brokerages/research institutions through a four-stage pipeline (outline → data collection → parallel writing → review and assembly). Cost is less than 0.6 yuan, takes 10–20 minutes, supports output in 19 languages, suitable for independent developers and researchers.
Researchers from HKUST, ByteDance, and UCL propose SCORE, a co-evolutionary training framework that jointly trains an LLM as both a deep research report generator and an evaluator, using a meta-harness to dynamically adjust evaluation difficulty and prevent reward saturation. Experiments show consistent improvement in open-ended research report quality.
Introduces TVIR, a benchmark and hierarchical multi-agent framework for generating text-visual interleaved reports, evaluating factual reliability and visual alignment in automated report generation.
This paper presents Ptah, a multi-agent harness for generating verifiable multimodal deep research reports by interleaving textual and visual evidence through specialized agents and verification mechanisms. It introduces PtahEval for evaluation.
The article demonstrates how to use Sense Nova Skills, an AI tool, to generate a full global EV industry research report from a single prompt, with links to the GitHub repo and plugin.
AnchorDiff proposes a topology-aware masked diffusion framework for radiology report generation, integrating RadGraph-derived clinical anchors and confidence-based rewriting to achieve state-of-the-art results on MIMIC-CXR and MIMIC-RG4 benchmarks.
Google DeepMind launched Deep Research and Deep Research Max, autonomous agents using Gemini 3.1 Pro to browse web and custom data for professional, fully-cited reports.
DR³-Eval is a benchmark for evaluating deep research agents on multimodal, multi-file report generation with a realistic web environment simulation and comprehensive evaluation framework measuring information recall, factual accuracy, citation coverage, instruction following, and depth quality.