AI Paper Atlas

Research map / Video / Video-language understanding

GranAlign: Granularity-Aware Alignment Framework for Zero-shot Video Moment Retrieval

Mingyu Jeon, Sunjae Yoon, Jonghee Kim, Junyeong Kim

AAAI 2026 · Video › Video-language understanding · approach: Video moment retrieval · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use query-aware caption instead of query-agnostic caption with raw query (ablation)GranAlign · QVHighlights valmAP@avg: 31.8 -> 32.13Improved
pair simplified query with query-aware caption (granularity mismatch) (ablation)GranAlign · QVHighlights valmAP@avg: 37.13 -> 36.22Hurt
pair detailed query with query-aware caption instead of query-agnostic (ablation)GranAlign · QVHighlights valmAP@avg: 36.82 -> 37.65Improved
combine simplified-agnostic and detailed-aware pairs versus best single pair (ablation)GranAlign · QVHighlights valmAP@avg: 37.65 -> 39.12Improved
combine both pairs versus best single pair (ablation)GranAlign · QVHighlights valR1@0.5: 59.48 -> 61.94Improved
use simplified query with query-agnostic caption instead of raw query (ablation)GranAlign · QVHighlights valmAP@avg: 31.8 -> 37.13Improved
swap granularity assignment (detailed with agnostic, simplified with aware) (ablation)GranAlign · QVHighlights valR1@0.5: 61.94 -> 60.90Hurt
use detailed queries for both caption types instead of granularity-matched pairing (ablation)GranAlign · QVHighlights valmAP@avg: 39.12 -> 39.01Hurt
use three rewritten queries and weight 0.3 versus nearby values (ablation)GranAlign · QVHighlights valperformanceNo clear effect
GranAlign versus Moment-GPT on QVHighlights valGranAlign · QVHighlights valmAP@avg: 35.9 -> 39.12Improved
GranAlign versus Moment-GPT on ActivityNet-CaptionsGranAlign · ActivityNet-CaptionsR1@0.5: 31.1 -> 34.0Improved
GranAlign versus Moment-GPT on Charades-STAGranAlign · Charades-STAR1@0.5: 38.4 -> 39.6Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy