VLAN - (c) Wheelchair - Semantic Success: leaderboard

Full1931trajectorysplit; user-conditionedlanguageinstruction. SRlan requiresalluser-specificsemanticconstraints; IAS weightedstructuredsimilaritytocanonicaltarget. Seven releasedVLMs; exactInstructlabelsretained; noactionSRderivedmetricused. Metric: Native score (0–1). Source: https://arxiv.org/html/2611.18346v1

Metric: Native score (0–1). Source: arxiv.org. Saturation forecast: Around September 2028. 7 models tracked. Updated 17 Nov 2026.

Top models

#ModelScoreOverall rank
1InternVL3.5-8B0.347#1588
2Qwen 2.5 VL 3B Instruct0.332#2648
3Molmo2-8B0.289#1629
4Qwen 2.5 VL 7B Instruct0.208#1893
5Qwen 3 VL 8B Instruct0.092#1025
6Llama 3.2 11B Instruct0.075#5226
7Idefics3-8B-Llama30.004#5775

No result here: #1 Claude Fable 5.2, #2 GPT-6.2 Astra, #3 Gemini 4 Carbon.

Interactive version: theaggregate.ai/benchmark?slug=vlan-c-wheelchair-semantic-success · How It Works · Data refreshed daily, snapshot 2026-11-17.