
IJ
Ishani Janveja, Davis Zhang, Seoyul Oh, Deepak Vasisht
· 1 min read
ResearcharXiv cs.CV
Exploiting answer-invariant redundancies in satellite imagery for efficient VLM inference on edge
arXiv:2609.29029v1 Announce Type: new
Abstract: Onboard vision-language models could enable satellites to answer queries directly, but exhaustive tiled inference over high-resolution imagery is slow and energy-intensive. We identify answer-invariant token redundancy (AITR): image tiles and vision tokens that can be removed without changing the final answer. We present Rift, a two-stage system that performs query-conditioned tile pruning followed by elastic prefill to reduce token budget. We evaluate it on LLaVA-1.5 7B running on Jetson AGX Orin. Compared with exhaustive tiled inference, Rift reduces energy by 78% and latency by 69%, while increasing accuracy from 45% to 73%.
Original source
This story was published by arXiv cs.CV and written by Ishani Janveja, Davis Zhang, Seoyul Oh, Deepak Vasisht. SyncAI.news shows a preview; the complete article is on the publisher's site.
Read the full story on arxiv.org


