Multimodal Generation on M2LONGBENCH (test)
39.7Anchor DescriptionDEEP-REPORTER
Evaluation results
| Method | Links | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| DEEP-REPORTERBackbone=Qwen3-32B, Training Strategy=SFT, Retrieval Filtering=true, Recurrent Context Management=true2026.04 | 39.7 | 39.7 | 39.7 | 47.8 | 37.2 | 38.5 | 43.3 | 41.7 | 27.1 | 36.4 | 36.8 | 28.7 | 32.3 | 37.9 | 4,700 | |
| DEEP-REPORTERBackbone=Qwen3-32B, Retrieval Filtering=true2026.04 | 38.1 | 38.9 | 38.5 | 12.2 | 10.5 | 15.4 | 10.5 | 12.2 | 24.3 | 35.2 | 34.8 | 29.2 | 30.9 | 27.2 | 3,900 | |
| DEEP-REPORTERBackbone=Qwen3-32B, Training Strategy=SFT, Recurrent Context Management=false2026.04 | 36.4 | 38.8 | 37.6 | 39.2 | 30.4 | 32.8 | 34.4 | 34.2 | 11.3 | 24.3 | 25 | 18.6 | 19.8 | 30.5 | 4,500 | |
| DEEP-REPORTERBackbone=Qwen3-32B, Training Strategy=SFT, Retrieval Filtering=false, Recurrent Context Management=false2026.04 | 36 | 37.8 | 36.9 | 27.5 | 20.2 | 25.2 | 25.1 | 24.5 | 9.7 | 21.5 | 22.6 | 15.8 | 17.4 | 26.3 | 4,300 | |
| DEEP-REPORTERBackbone=Llama3.3-70B, Training Strategy=SFT2026.04 | 35.6 | 37.7 | 36.6 | 36.4 | 36.4 | 37.7 | 36.4 | 36.7 | 33.2 | 32.8 | 30.4 | 27.5 | 31 | 34.8 | 4,500 | |
| DEEP-REPORTERBackbone=Qwen3-32B, Training Strategy=SFT, Retrieval Filtering=false2026.04 | 34.8 | 38.1 | 36.4 | 30.4 | 24.3 | 28.7 | 27.1 | 27.6 | 23.5 | 25.1 | 23.9 | 25.9 | 24.6 | 29.6 | 4,900 | |
| DEEP-REPORTERBackbone=Qwen3-32B, Retrieval Filtering=false2026.04 | 31.6 | 36.8 | 34.2 | 9.7 | 4.9 | 7.7 | 5.7 | 7 | 19.8 | 30.8 | 34 | 30.4 | 28.7 | 23.3 | 4,000 | |
| DEEP-REPORTERBackbone=Qwen3-8B, Training Strategy=SFT, Retrieval Filtering=false2026.04 | 24.7 | 27.9 | 26.3 | 30 | 21.5 | 27.1 | 28.3 | 26.7 | 15 | 25.9 | 23.9 | 21.1 | 21.5 | 24.8 | 5,200 | |
| DEEP-REPORTERBackbone=Qwen3-8B, Training Strategy=SFT, Retrieval Filtering=true2026.04 | 23.9 | 26.3 | 25.1 | 44.1 | 36.4 | 40.1 | 43.7 | 41.1 | 17.8 | 23.9 | 22.7 | 19 | 20.9 | 29 | 5,100 | |
| Storm-MMBackbone=Qwen3-32B2026.04 | 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 | 3,200 | |
| DEEP-REPORTERBackbone=Qwen3-32B, Training Strategy=DPO2026.04 | 19.4 | 20.2 | 19.8 | 15.8 | 8.5 | 12.2 | 10.1 | 11.6 | 14.2 | 28.3 | 27.9 | 23.9 | 23.6 | 18.4 | 3,900 | |
| DEEP-REPORTERBackbone=Qwen3-8B, Training Strategy=DPO, Retrieval Filtering=false2026.04 | 17.4 | 23.5 | 20.4 | 8.1 | 3.6 | 6.5 | 2.4 | 5.2 | 10.9 | 22.7 | 16.6 | 21.9 | 18 | 14.5 | 4,100 | |
| DEEP-REPORTERBackbone=Qwen3-8B, Training Strategy=DPO, Retrieval Filtering=true2026.04 | 16.6 | 16.6 | 16.6 | 9.3 | 5.3 | 7.3 | 6.5 | 7.1 | 12.6 | 25.5 | 21.1 | 19.8 | 19.7 | 14.5 | 4,200 | |
| DEEP-REPORTERBackbone=Qwen3-8B, Retrieval Filtering=true2026.04 | 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 | 3,900 | |
| DEEP-REPORTERBackbone=Qwen3-8B, Retrieval Filtering=false2026.04 | 13 | 14.6 | 13.8 | 8.1 | 2 | 6.1 | 3.2 | 4.9 | 8.1 | 21.9 | 17.8 | 20.7 | 17.1 | 11.9 | 3,800 | |
| DEEP-REPORTERBackbone=Llama3.3-70B, Training Strategy=DPO2026.04 | 4.9 | 7.3 | 6.1 | 11.7 | 6.5 | 10.5 | 8.9 | 9.4 | 4.9 | 13.8 | 9.3 | 7.3 | 8.8 | 8.1 | 3,500 | |
| DEEP-REPORTERBackbone=Llama3.3-70B2026.04 | 4.1 | 6.5 | 5.3 | 11.3 | 6.1 | 9.3 | 6.5 | 8.3 | 4.5 | 8.1 | 10.5 | 5.7 | 7.2 | 6.9 | 3,400 | |
| Naïve RAGBackbone=Qwen3-8B2026.04 | 0 | 2 | 1 | 6.1 | 7.9 | 7.1 | 8.5 | 7.4 | 0 | 0 | 6.9 | 23.1 | 7.5 | 5.3 | 1,900 | |
| Naïve RAGBackbone=Qwen3-32B2026.04 | 0 | 1.2 | 0.6 | 6.9 | 5.7 | 6.3 | 6.7 | 6.4 | 0 | 0 | 10.1 | 24.3 | 8.6 | 5.2 | 2,000 |