AI Paper Atlas

Research map / Video / Video-language understanding

Listening Between the Frames: Bridging Temporal Gaps in Large Audio-Language Models

Hualei Wang, Yiming Li, Shuo Ma, Hong Liu, Xiangdong Wang

AAAI 2026 · Video › Video-language understanding · approach: Video LLMs with fine-grained understanding · 12 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
add temporal marker tokens with numeral-based init (ablation)SALMONN FTAR-tuned · AudioSet-Strong dense captioningEb-F1: 32.4 -> 36.0Improved
initialize temporal marker embeddings from numeral token embeddings instead of random (ablation)temporal markers · AudioSet-Strong dense captioningEb-F1: 33.5 -> 36.0Improved
add absolute time-aware encoding to temporal markers (ablation)SALMONN + TM · AudioGroundingmIoU: 54.9 -> 56.0Improved
add absolute time-aware encoding only (ablation)SALMONN FTAR-tuned · AudioGroundingmIoU: 51.9 -> 53.8Improved
add segment-level token merging (ablation)SALMONN + TM + ATE · CNN/DailyMail speech timelinemIoU: 90.4 -> 94.2Improved
add segment-level token merging (ablation)SALMONN + TM + ATE · AudioSet-Strong dense captioningEb-F1: 37.8 -> 37.4Hurt
add segment-level token merging (ablation)SALMONN + TM + ATE · AudioGroundingmIoU: 56.0 -> 57.8Improved
raise retained token ratio from 0.25 to 0.30 (ablation)TimeAudio · CNN/DailyMail speechROUGE-1: 42.4 -> 42.5No clear effect
lower retained token ratio from 0.25 to 0.10 (ablation)TimeAudio · CNN/DailyMail speechROUGE-1: 42.4 -> 24.3Hurt
TimeAudio vs fine-tuned Qwen2-AudioTimeAudio · AudioSet-Strong dense captioningMETEOR: 22.4 -> 20.4Hurt
TimeAudio vs FTAR-tuned SALMONN-7BTimeAudio · AudioGroundingmIoU: 51.9 -> 57.8Improved

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