Search papers, labs, and topics across Lattice.
This paper introduces Aphanta, a diagnostic framework that evaluates the effectiveness of image editors in enhancing multimodal large language models (MLLMs) by assessing their performance across various task conditions. The study reveals that the utility of image editing is highly task-dependent, with significant improvements observed in tasks involving visual cue injection and grounding, while tasks requiring complex symbolic manipulation showed poor reliability. The consolidated Qwen pipeline demonstrates a notable increase in mean task scores, highlighting the potential of image editing as a specialized tool rather than a one-size-fits-all solution for multimodal reasoning.
Task-specific image editing can boost MLLM performance by over 29%, but not all tasks benefit equally鈥攕ome are left behind.
Explicit visual intermediates can help multimodal large language models (MLLMs) externalize spatial evidence and updated visual states, but their utility depends on whether an image editor can faithfully realize the required transformation. We introduce \textbf{Aphanta}, an automated task-discovery and closed-loop diagnostic framework for the MLLM ->image editor ->MLLM pipeline. Aphanta evaluates three conditions---direct reasoning, reasoning with an editor-generated intermediate, and reasoning with an idealized reference intermediate---to separate potential visual headroom from the practical utility of current editors. Across 20 candidate tasks and multiple editor--MLLM combinations, we find that utility is strongly task-conditioned. Gains concentrate in visual cue injection, grounding, and counterfactual state realization, whereas intermediates requiring symbol-sensitive construction or structural extrapolation are substantially less reliable. On the selected positive-task subset, our consolidated Qwen pipeline improves the mean task score from 0.343 to 0.445 ($+10.2$ points; $+29.7\%$ relative), while the full study also retains filtered and unsuccessful tasks to expose the boundary. These results position image editing as a specialized visual workspace rather than a universal reasoning mechanism, and establish Aphanta as a reusable protocol for measuring task--representation alignment, editor realization, and downstream pipeline utility.
