ARCANA is a new benchmark designed to test far analogical reasoning, requiring mappings from real-world objects or phenomena to abstract grid-based transformations. Human participants reliably solve these tasks, whereas large language models exhibit a significant performance gap, highlighting systematic failure modes. The authors introduce ConceptSIR, a framework that improves LLM performance on challenging ARCANA tasks, yet far analogical reasoning remains an open problem for current models.
By Ruqi Chen, Zilei Wang, Wei Wang
The paper introduces a two‑step test‑time training protocol, Embed‑TTT, for Vision ARC (VARC) that first fine‑tunes only the task embedding and then fine‑tunes the backbone. This approach consistently produces task embeddings that better align with the underlying rules, improves retrieval and linear probing, and recovers the geometric structure of parametric rules. Even fine‑tuning only the tiny embedding component solves a significant portion of ARC‑AGI‑1, ConceptARC, and Mini‑ARC tasks, while the full two‑step pipeline further enhances performance and demonstrates compositional rule interpolation.
By Adrien Deli\`ege, Claas Beger, Marc Van Droogenbroeck, Melanie Mitchell
PotARCin expands the ARC benchmark by evaluating abstract reasoning across five dimensions—Definition, Classification, Constrained Generation, Editing, and Inversion—using programmatic generation of new task instances. The study shows a 25‑52 percentage‑point performance gap between standard ARC evaluation and PotARCin, and reveals that multi‑dimensional assessment can reorder models that appear equivalent under single‑metric accuracy. Additionally, a new held‑out set, P‑ARC, demonstrates low model accuracy (1‑8%) across all dimensions, highlighting the need for more comprehensive tests of abstract reasoning.
By Claas Beger, Ryan Yi, Melanie Mitchell
ClueTracer is a training‑free, parameter‑free plugin that suppresses hallucinations in multimodal reasoning models by tracing how visual clues propagate from the question through the model’s reasoning pathway. It localizes task‑relevant image patches while ignoring irrelevant regions, thereby addressing the problem of reasoning drift where models over‑focus on question‑irrelevant entities. The method improves performance across various reasoning architectures by an average of 1.21× on reasoning benchmarks.
By Gongli Xi, Kun Wang, Zeming Gao, Huahui Yi, Haolang Lu, Qianshan Wei, Ye Tian, Wendong Wang
arXiv:2605.11388v2 Announce Type: replace
Abstract: Humans intuitively solve complex problems by flexibly shifting among reasoning modes: they plan, execute, revise intermediate goals, resolve ambigu...
By Dean Light, Michael Theologitis, Kshitish Ghate, Shuyue Stella Li, Benjamin Newman, Chirag Shah, Aylin Caliskan, Pang Wei Koh, Dan Suciu, Yulia Tsvetkov
arXiv:2511. 18121v2 Announce Type: replace-cross Abstract: While Multimodal Large Language Models (MLLMs) excel on benchmarks, their processing paradigm differs from the human ability to integrate visual information.
By Ming Zhong, Yuanlei Wang, Liuzhou Zhang, Ruichuan An, Renrui Zhang, Hao Liang, Ming Lu, Ying Shen, Wentao Zhang