Deep-Reporter: Deep Research for Grounded Multimodal Long-Form Generation

Fangda Ye; Zhifei Xie; Yuxin Hu; Yihang Yin; Shurui Huang; Shikai Dong; Jianzhu Bao; Shuicheng Yan

cs.CLApr 12, 2026

Deep-Reporter: Deep Research for Grounded Multimodal Long-Form Generation

Authors: Fangda Ye, Zhifei Xie, Yuxin Hu, Yihang Yin, Shurui Huang, Shikai Dong, Jianzhu Bao, Shuicheng Yan

Organizations: National University of Singapore · Nanyang Technological University · University of Edinburgh · Beijing Institute of Technology

Abstract

Recent agentic search frameworks enable deep research via iterative planning and retrieval, reducing hallucinations and enhancing factual grounding. However, they remain text-centric, overlooking the multimodal evidence that characterizes real-world expert reports. We introduce a pressing task: multimodal long-form generation. Accordingly, we propose Deep-Reporter, a unified agentic framework for grounded multimodal long-form generation. It orchestrates: (i) Agentic Multimodal Search and Filtering to retrieve and filter textual passages and information-dense visuals; (ii) Checklist-Guided Incremental Synthesis to ensure coherent image-text integration and optimal citation placement; and (iii) Recurrent Context Management to balance long-range coherence with local fluency. We develop a rigorous curation pipeline producing 8K high-quality agentic traces for model optimization. We further introduce M2LongBench, a comprehensive testbed comprising 247 research tasks across 9 domains and a stable multimodal sandbox. Extensive experiments demonstrate that long-form multimodal generation is a challenging task, especially in multimodal selection and integration, and effective post-training can bridge the gap.

Figures & tables

Figure 1: Comparison of paradigms in long-form report generation. While traditional systems (a) lack visual engagement and pure T2I approaches (b) suffer from factual hallucinations and narrative fragmentation, our method (c) achieves high coherence and factuality by retrieving and integrating real-world visual evidence.

Figure 2: Deep-Reporter Architecture. (a) The multi-agent framework orchestrates planning, multimodal information seeking, and incremental writing to generate professional reports. (b) The data synthesis pipeline constructs high-quality expert trajectories to equip open-weight models with deep research capabilities.

Panel of Table 1 — Table 1: M 2 LongBench versus existing benchmarks. Evi. denotes candidate sets for text/image retrieval. Modality describe the expected input/output format (MM=Multimodal).

Method	Backbone	Search			Filter			Selection
Method	Backbone	Text	Image	Overall	Text	Image	Overall	Text	Image	Overall
Naïve RAG	Qwen3-8B	34.9 (54.6)	32.7 (27.3)	33.8 (81.9)	-	-	-	46.3 (18.7)	7.8 (0.9)	27.0 (19.6)
Naïve RAG	Qwen3-32B	35.0 (54.7)	32.8 (27.4)	33.9 (82.0)	-	-	-	45.7 (21.0)	10.4 (1.0)	28.1 (22.0)
Storm-MM	Qwen3-32B	29.5 (105.2)	38.4 (53.8)	34.0 (159.0)	-	-	-	48.3 (26.3)	12.7 (1.4)	30.5 (27.7)
Deep-Reporter	Qwen3-8B	29.3 (103.8)	37.9 (52.4)	33.6 (156.3)	45.5 (50.4)	50.2 (30.5)	47.9 (81.0)	49.6 (24.5)	8.3 (0.7)	28.9 (25.2)
w/o Filter	Qwen3-8B	29.5 (104.0)	37.2 (52.5)	33.4 (156.5)	-	-	-	43.6 (31.9)	3.8 (0.3)	23.7 (32.2)

Table 2: Evidence quality on M 2 LongBench benchmark. Each cell shows precision (%) and quantity of evidence that are either retrieved/retained/cited. The best score is in purple boldface and second best is underlined . ♠ and ♣ denote SFT and DPO respectively. w/o Filter removes the filter module; w/o Recur. removes recurrent context management; w/o Both removes both.

Method	Backbone	Section Anchor			Section Content					Full Report					Overall	Report
Method	Backbone	Desc	Check	Avg	Rich	Coh	Place	Clar	Avg	Coh	Flu	Rep	Term	Avg	Overall	Length
Naïve RAG	Qwen3-8B	0.0	2.0	1.0	6.1	7.9	7.1	8.5	7.4	0.0	0.0	6.9	23.1	7.5	5.3	1.9k
Naïve RAG	Qwen3-32B	0.0	1.2	0.6	6.9	5.7	6.3	6.7	6.4	0.0	0.0	10.1	24.3	8.6	5.2	2.0k
Storm-MM	Qwen3-32B	22.3	25.3	23.8	7.7	5.7	7.3	6.5	6.8	10.1	20.7	19.4	14.6	16.2	15.6	3.2k
Deep-Reporter	Qwen3-8B	14.2	19.4	16.8	8.1	3.6	6.9	3.6	5.6	10.5	23.5	21.9	18.6	18.6	13.7	3.9k
w/o Filter	Qwen3-8B	13.0	14.6	13.8	8.1	2.0	6.1	3.2	4.9	8.1	21.9	17.8	20.7	17.1	11.9	3.8k

Table 3: Main results of generation quality. Section Anchor : Desc=Description, Check=Checklist. Section Content : Rich=Richness, Coh=Image-text coherence, Place=Placement, Clar=Clarity. Full Report : Coh=Coherence, Flu=Fluency, Rep=Repetition, Term=Termination. ♠ and ♣ denote SFT and DPO respectively. w/o Recur. removes recurrent context management; w/o Both removes both filtering and recurrent context.

Figure 3: Overall performance with output tokens.

Appendix figures & tables8 assets

Supplementary material from the paper’s appendix.

Appendix

Figure 4: Demonstration of the sandbox construction.

Domain	Source Report Metadata			Tasks		Sandbox Scale (Original / Expanded)				Silver Annot.
Domain	Count	Image	Token	Sec.	Check.	Web Pages	PDF Docs	Images	Chunks	Images	Chunks
AI	43	12.2	7.3k	5.7	3.6	28.5 / 59.8	16.8/ 27.9	39.7 / 289.9	103.0k / 179.3k	98.7	170.2
Finance	37	10.8	7.5k	5.9	3.4	16.4 / 79.5	1.7/ 13.3	20.8 / 285.9	59.3k / 238.2k	110.5	176.0
Geography	12	8.8	4.1k	5.6	3.0	8.3 / 80.1	2.0 / 9.5	45.6 / 548.3	29.9k / 240.0k	99.7	165.3
Health	28	7.0	4.2k	5.8	3.3	19.2 / 77.0	5.5 / 6.7	45.3 / 287.5	70.0k / 230.6k	108.5	172.5
History	22	7.4	3.3k	5.1	3.1	3.3 / 59.0	1.5 / 11.1	8.3 / 592.6	23.0k / 324.1k	99.1	154.0

Appendix

Table 4: Statistical overview of M 2 LongBench . Source Report Metadata reflects the complexity of the ground-truth blueprints. Sandbox Scale highlights the difficulty of the retrieval task by contrasting original citations with our significantly expanded search space (Expanded), demonstrating the “needle-in-a-haystack” nature of the benchmark.

Model	Revision (Git commit SHA)
Qwen3-8B	b968826d9c46dd6066d109eabc6255188de91218
Qwen3-32B	9216db5781bf21249d130ec9da846c4624c16137
Llama-3.3-70B-Instruct	6f6073b423013f6a7d4d9f39144961bfbfbc386b
InternVL3-5-38B	de99855be3642cd44fe97c9b72d70e5ce2c07f69
Jina-Embeddings-v4	5f4b9cbb80cc95ba44fe6667dfd75710f7db2947

Appendix

Table 5: Model versions and Hugging Face Hub revisions used in our experiments.

Parameter	Value
Training Objective	Next-token prediction
Optimizer	AdamW ( $\beta_{1}$ =0.9, $\beta_{2}$ =0.95)
Learning Rate	1e-4
LR Scheduler	Cosine decay
Warmup Ratio	0.05
Global Batch Size	64

Appendix

Table 6: Supervised Fine-Tuning (SFT) hyperparameters.

Parameter	Value
Loss Type	DPO (sigmoid)
Beta ( $\beta$ )	0.1
Learning Rate	5e-6
LR Scheduler	Cosine decay
Warmup Ratio	0.05
Global Batch Size	64

Appendix

Table 7: Direct Preference Optimization (DPO) configuration.

Model	Filter	Train	Total	Query	Search	Filter	Write
Qwen3-8B	✓	Base	630	10.6	30.4	541.4	47.5
Qwen3-8B	✗	Base	170	18.1	75.1	-	77.2
Qwen3-8B	✓	SFT	636	12.1	31.6	530.3	61.5
Qwen3-32B	✓	Base	644	28.1	29.0	434.3	152.6
Qwen3-32B	✗	Base	342	43.7	38.4	-	259.6
Qwen3-32B	✓	SFT	625	37.9	31.8	369.1	186.7

Appendix

Table 8: End-to-end runtime breakdown (seconds) averaged over 247 tasks. Query : LLM generates search queries. Search : retrieval service latency. Filter : LLM filters retrieved results. Write : LLM generates section content.

Figure 5: Model performance evaluation. (A-C) Overall comparisons: comprehensive metrics, retrieval pipeline stages, and detailed quality indicators. (D-I) Subject-specific retrieval performance for image and text modalities across different pipeline stages.

Figure 6: Subject-specific generation quality score. (J-O) Article and section quality across six dimensions: coherence, fluency, repetition control, termination, description completeness, and checklist adherence. (P-T) Citation quality across five dimensions: overall performance, richness, coherence, placement, and clarity. Performance patterns vary significantly across academic domains.

Multimodal Generation

Multimodal generation creates coordinated text, images, audio, video, 3D, and other outputs from mixed inputs while preserving meaning and consistency.

See all 243 papers

Towards Verifiable Multimodal Deep Research: A Multi-Agent Harness for Interleaved Report Generation
May 28, 2026Chenghao Zhang, Guanting Dong, Yufan Liu +3Multimodal Generation Multimodal Agents
Gaoling School of Artificial Intelligence, Renmin University of China
MIRA: A Multimodal Illustrated Deep Research Agent for Controllable Reports
Oct 19, 2026Hongrui Wu, Kaiwen Tuo, Congcong Wang +7Deep Research Agents Multimodal Agents
Stanford University · The Hong Kong University of Science and Technology · Fudan University +2
MM-DeepResearch: A Simple and Effective Multimodal Agentic Search Baseline
Mar 1, 2026Huanjin Yao, Qixiang Yin, Min Yang +5Multimodal Search Agents Multimodal QA
ByteDance · Zhongguancun Academy · Nanyang Technological University +2

May 28, 2026J/K move · Enter open · S save

Towards Verifiable Multimodal Deep Research: A Multi-Agent Harness for Interleaved Report Generation

Chenghao Zhang, Guanting Dong, Yufan Liu, Tong Zhao+2

Gaoling School of Artificial Intelligence, Renmin University of China

Multimodal Generation Multimodal Agents Multi-Agent LLMs Multimodal Model Evaluation Multimodal Tool Use

Open paper AskPDF arXiv

Large Language Models (LLMs) have advanced autonomous agents from deep search, which retrieves concise factual answers, to deep research, which synthesizes scattered evidence into long-form reports. However, verifiable multimodal deep research remains challenging due to open-ended synthesis without deterministic ground truth and the need to interleave textual arguments with visual evidence. We propose Ptah, a multi-agent harness for interleaved report generation. Ptah orchestrates the lifecycle from user query to rendered web report through planning, research, and writing stages, where specialized agents construct visual-aware plans, collect claim-grounded evidence, maintain source-aligned images in a Visual Working Memory, and compose reports through declarative multimodal tool use. A verifier agent serves as the harness's acceptance function, enforcing factual grounding, citation fidelity, and cross-modal consistency throughout the workflow. We further introduce PtahEval, an evaluation protocol that augments existing benchmarks with image-level and presentation-level assessments. Experiments on deep research benchmarks show that Ptah produces more reliable, visually informative, and usable human-facing multimodal reports than strong baselines. Our code is released at https://github.com/SnowNation101/Ptah

Abstract

Multimodal Generation

Explore similar work

Towards Verifiable Multimodal Deep Research: A Multi-Agent Harness for Interleaved Report Generation

MIRA: A Multimodal Illustrated Deep Research Agent for Controllable Reports

MM-DeepResearch: A Simple and Effective Multimodal Agentic Search Baseline