AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Stop Mixing Things Up! BISCUIT Teaches Vision-Language Models to Learn New Concepts from Images on the Spot

Jiahua Bao, Siyao Cheng, Jiaxing Du, Yuhang Jia, Boyang Niu, Zeming Lang, Changjiang He, Hao Zhang, Jie Liu

AAAI 2026 · Vision-language and multimodal › Multimodal LLMs · approach: VLM capability analysis · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes BISCUIT, a method that teaches vision-language models to learn and use newly introduced visual concepts from images at inference time, addressing the "Mixing Things Up" failure where concept names get confused.

Problem

VLMs often fail to apply newly introduced visual concepts during inference, confusing concept names and giving vague descriptions and poor grounding; prior work mainly handled person-related concepts through text prompts.

Approach

Teaches VLMs to learn new visual concepts directly from images on the spot, rather than relying on text prompts designed mainly for person-related concepts (details beyond the abstract not available).

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add BISCUIT trainingQwen2.5-VL-7B · BiscuitVQA overallSANLS summed: 339.9 -> 483.9Improved
add BISCUIT trainingLLaVA-1.5-7B · BiscuitVQA overallSANLS summed: 274.0 -> 325.2Improved
add dual-stream (RGB + edge) encoder (ablation)BISCUIT on Qwen2.5-VL-7B · BiscuitVQA overallSGPT summed: 731.4 -> 799.6Improved
add SVD-based initialization of edge-stream modules (ablation)BISCUIT on Qwen2.5-VL-7B · BiscuitVQA overallSGPT summed: 738.6 -> 799.6Improved
add identification-aware loss instead of standard DPO (ablation)BISCUIT on Qwen2.5-VL-7B · BiscuitVQA overallSGPT summed: 721.7 -> 799.6Improved
add identification-aware loss instead of standard DPO (ablation)BISCUIT on Qwen2.5-VL-7B · Generalization retentionretention (%): 94.7% -> 94.1%No clear effect
use Canny edges instead of no edge detection (ablation)BISCUIT on Qwen2.5-VL-7B · IICLIICL: 608.0 -> 641.8Improved
use Sobel edges instead of Canny (ablation)BISCUIT on Qwen2.5-VL-7B · IICLIICL: 641.8 -> 587.4Hurt
use QR basis fusion instead of direct sum (ablation)BISCUIT on Qwen2.5-VL-7B · IICLIICL: 621.5 -> 641.8Improved
set threshold tau=0.0 instead of 0.1 (ablation)identification-aware loss · Generalization, Qwen2.5-VL-3BgeneralizationHurt
use BISCUIT vs other concept-learning methods on generalizationQwen2.5-VL-7B · Generalization retention across 11 benchmarksretention (%): 91.09% -> 94.07%Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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