Long-document visual question answering (VQA) over documents of tens to hundreds of pages mixing text, tables, charts, and figures typically follows retrieve-then-read pipelines. In our setting, the bottleneck shifts from retrieval recall to reranker-side evidence selection: on MMLongBench-Doc, BGE-M3 reaches Recall@20 = 0.86 but only F1@5 = 0.254, and even the visual retriever ColPali reaches only F1@5 = 0.332; a text-only rerank LLM seeing only raw snippets misses table, chart, and layout evidence even when the upstream retriever encoded images. We propose Trident, with two complementary components: Trident-R, a retriever-agnostic LLM reranker that converts each candidate into an LLM-readable semantic record, including a visual caption, section path, entity tags, multi-axis concept hits, and a text snippet, then performs a single adaptive-K rerank call; and Trident-S, a generation-side module that prompts the VLM under topical, entity, and structural lenses before synthesis. On two long-document datasets, the annotation+rerank protocol substantially improves retrieval F1 across five heterogeneous pools, with every reranked pool exceeding the strongest adaptive-K baseline PageIndex. An LLM rerank without the annotation barely changes first-hit ranking, indicating the lift comes from the structured annotation. Trident-S targets open-ended synthesis questions by design, adding up to 6.6 points in generation accuracy on these questions. The best Trident configuration is the strongest downstream QA pipeline in our evaluation, with rankings consistent across two LLM judges (kappa = 0.913).
Figures & tables
Figure 1: Overview of Trident . Given a query and a long document, a retriever first returns a top-20 candidate pool where evidence pages often already appear. Trident-R improves evidence selection by converting candidates into semantic records for adaptive LLM reranking, while Trident-S synthesizes the final answer through topical, entity, and structural lenses.
Figure 2: Overall framework of TRIDENT-R.
Method
avg. K
Recall
Prec.
F1
MRR
nDCG
Fixed- K baselines (no rerank, K=5 )
BM25
5.00
0.525
0.162
0.231
0.448
0.437
RRF
5.00
0.556
0.172
0.246
0.524
0.496
BGE-M3
5.00
0.572
0.178
0.254
0.524
0.502
ColPali
5.00
0.739
0.236
0.332
0.692
0.675
Adaptive- K , single-aspect signal
Table 1: Cross-pool isolation on MMLongBench-Doc ( N=830 aligned QAs; all reranked rows use GPT-4.1 with Kmax=5 ). The same annotation + rerank protocol improves sparse, hybrid, graph, dense, and visual candidate pools.
Pool
Candidate view
F1
MRR
BGE-M3
raw top-5 retrieval
0.254
0.524
BGE-M3
text snippet + LLM rerank
0.374
0.523
BGE-M3
caption-only annotation + rerank
0.518
0.692
BGE-M3
full annotation + rerank
0.546
0.744
ColPali
raw top-5 retrieval
0.332
0.692
ColPali
text snippet + LLM rerank
0.423
0.598
Table 2: Mechanism ablations. LLM reranking with only raw text snippets does not improve first-hit ranking on either pool (MRR flat on BGE-M3, and decreasing on ColPali because the LLM judgment overrides ColPali’s visual ranking signal); the ranking lift requires the structured annotation, and the visual caption alone recovers most of it on both text and visual pools.
Method
GPT J.
Claude J.
Gen Acc
Gen F1
BGE-M3 (Plain)
44.7
44.7
43.5
41.3
ColPali (Plain)
48.3
46.3
46.4
45.9
PageIndex + Trident-S
49.3
48.3
45.9
42.6
BGE-M3 + ann. + rk + Trident-S
52.0
50.0
48.9
46.4
ColPali + ann. + rk, Plain
51.0
49.7
47.1
44.5
ColPali + ann. + rk + Trident-S
53.7
52.7
51.2
48.9
Table 3: End-to-end QA on the 300-QA MMLongBench-Doc subset (selected pipelines; all values in %). Full 14-pipeline results, including answer accuracy and unanswerable abstention, are in Appendix A .
Method
avg. K
Recall
Prec.
F1
MRR
nDCG
Fixed- K baselines (no rerank, K=5 )
BM25
5.00
0.623
0.191
0.281
0.616
0.568
BGE-M3
5.00
0.649
0.205
0.298
0.622
0.575
RRF
5.00
0.684
0.217
0.315
0.675
0.622
ColPali
5.00
0.750
0.235
0.343
0.763
0.700
Adaptive- K
Table 4: LongDocURL retrieval comparison ( N=1,122 QAs with non-empty gold evidence). The retrieval-side trend transfers, although the full cross-pool isolation study is conducted on MMLongBench-Doc.
Appendix figures & tables7 assets
Supplementary material from the paper’s appendix.
Appendix
Method
GPT-4.1 J.
Claude J.
Gen Acc
Gen F1
Ans. Acc
Unans. Abst.
BM25 (Plain)
42.3
42.7
41.3
38.2
33.7
65.3
BM25 + Trident-S
45.0
44.3
44.8
43.8
39.7
61.1
RRF (Plain)
44.3
44.3
42.6
39.6
36.3
62.5
RRF + Trident-S
44.7
43.3
43.9
41.7
39.3
58.3
BGE-M3 (Plain)
44.7
44.7
43.5
41.3
37.9
61.1
BGE-M3 + Trident-S
46.0
45.7
43.9
43.4
41.1
52.8
Appendix
Table 5: Full end-to-end QA results on the 300-QA MMLongBench-Doc subset (all values in %). J. =LLM-as-judge accuracy; + ann. =multi-aspect annotation; + rk =adaptive- K rerank; Plain =single multimodal VLM call; + Trident-S =three-lens generation with synthesis.
Method
GPT-4.1 J.
Claude J.
Gen Acc
Gen F1
Trident-R (Plain)
51.3
47.7
42.6
47.3
Trident-R + Trident-S
50.0
50.0
41.1
45.5
ColPali + ann. + rk, Plain
68.7
65.7
56.0
58.0
ColPali + ann. + rk, + Trident-S
66.3
66.0
56.7
58.5
Appendix
Table 6: End-to-end QA on the 300-QA LongDocURL stratified subset (all values in %; all questions are answerable). LongDocURL is dominated by MCQ-style and short-extractive answers, so Trident-S gives little average benefit.
Axis (category)
Full %
Subset %
Δ pp
Answer format
Float
14.8
20.7
+5.9
Int
26.4
24.0
−2.4
List
13.3
14.3
+1.0
Str
23.1
17.0
−6.1
None (unans.)
22.4
24.0
+1.6
Appendix
Table 7: Distribution match between the 300-QA subset and the full MMLongBench-Doc dataset. Most axes (Evidence Source, Evidence Pages, Unanswerable rate) match within ±3 pp. Answer format is moderately biased toward Float ( +5.9 pp) and away from Str ( −6.1 pp); document length is intentionally biased toward longer documents (median 72 vs. 28 pages), aligning with our framework’s long-document focus.
Table 9: Per-field annotation ablation on two retrievers ( N=830 aligned, Kmax=5 ). Standard rows: remove one field at a time. Caption-only rows : keep only the visual caption (drop concept_hits, section path, entity tags, text snippet). Caption alone achieves ∼90% of the full annotation lift on both pools ( −0.028 / −0.027 from full), with the gap remarkably pool-invariant. Combined with the asymmetric drop-caption result ( −0.130 on BGE-M3 vs. −0.083 on ColPali; 36% smaller on visual pool), this rules out pure modality completion as the sole mechanism: the lift stems from structuring page evidence as LLM-readable input, with visual caption acting as a semantic interface for the text-only LLM reranker (effective regardless of whether the upstream retriever is visual or textual).
Question type
BM25
BGE-M3
ColPali
Trident-R
+ ann
+ ann
+ ann
(xKG) + ann
Pure-text
2.95
3.08
3.11
2.94
Figure
2.64
2.61
2.60
2.71
Table
2.90
2.90
2.98
2.75
Chart
2.81
2.77
2.55
2.59
Layout
2.84
2.78
2.97
2.75
Appendix
Table 10: Mean adaptive K per question type, by candidate pool ( Kmax=5 ). Two patterns: (1) K is highly stable across pools (range 2.55 – 3.11 ), suggesting that the reranker’s K choice is driven by the question’s evidence requirement, not by which retriever fed the pool. (2) K varies by question type : Pure-text questions request the most pages ( Kˉ≈3.0 , K=5 cap rate ∼30% ); Figure / Chart questions request fewer ( Kˉ≈2.6 , K=1 rate >30% on Figure), consistent with visual evidence being concentrated on individual pages.
Pipeline
Agree.
Cohen κ
BM25 (Plain / MAVS)
94.0 / 96.7
0.89 / 0.94
BGE-M3 (Plain / MAVS)
94.3 / 97.3
0.90 / 0.95
ColPali (Plain / MAVS)
93.0 / 95.7
0.87 / 0.92
PageIndex (Plain / MAVS)
97.3 / 97.7
0.95 / 0.96
BGE-M3 + ann. (Plain / MAVS)
93.7 / 96.3
0.89 / 0.93
Trident-R (Plain / MAVS)
92.7 / 94.7
0.87 / 0.91
Appendix
Table 11: Inter-judge agreement and Cohen’s κ between GPT-4.1 and Claude on the same 300-QA outputs. Mean κ=0.913 corresponds to “almost perfect agreement” on the Landis-Koch scale. The pipeline-level ranking under the two judges is identical (ColPali + ann. + rk + Trident-S is best under both; Trident-R + Trident-S second under Claude, tied second under GPT). This indicates that the observed end-to-end ranking is not dependent on a single judge’s quirks.