ErrorRadar is a new benchmark of 2,500 multimodal K-12 math problems for MLLM error step identification and categorization, where GPT-4o trails human experts by ~10%.
hub
Visual sketchpad: Sketching as a visual chain of thought for multimodal language models
26 Pith papers cite this work. Polarity classification is still indexing.
26
Pith papers citing it
hub tools
citation-role summary
background 3
citation-polarity summary
roles
background 3polarities
background 3representative citing papers
Future-L1 interleaves latent visual spans with text in MLLM decoding, trained on a custom Future-L1-50K dataset via LA-DAPO RL, and reports SOTA gains on FutureBench (61.0 to 85.4) and TwiFF-Bench (2.44 to 3.04).
Sandboxed coding agents with text+image access match or outperform native omnimodal models on audio-video benchmarks by converting tasks into code-driven retrieval and processing.
Draw2Think recasts geometric reasoning as agentic interaction with a constraint engine, achieving 95.9% predicate-level construction fidelity and up to 16.4% accuracy gains on solid geometry tasks.
A planner-orchestrator system learns long-horizon image editing by maximizing outcome-based rewards from a vision-language judge and refining plans from successful trajectories.
DMLR performs dynamic visual-textual interleaving in latent space using confidence-guided latent policy gradient optimization and a dynamic visual injection strategy, yielding improved multimodal reasoning on benchmarks.
Latent Visual Reasoning enables autoregressive generation of latent visual states that reconstruct critical image tokens, yielding gains on perception-heavy VQA benchmarks such as 71.67% on MMVP.
PuzzleWorld benchmark reveals state-of-the-art AI models solve only 18% of complex puzzlehunt problems with 40% stepwise accuracy, matching novices but trailing enthusiasts, while fine-tuning on traces yields modest gains.
IPT supervision improves spatial reasoning in VLMs on perspective taking, path tracing, and multiview counting tasks, often outperforming textual chain-of-thought while remaining consistent with observed inputs.
VESTA introduces dynamic tool creation for VLMs that outperforms static-tool and no-tool baselines on distribution fitting, time series, and astronomy tasks in the new DAWN benchmark.
SeProD is a plug-and-play self-prophetic decoding framework that combines pre- and post-training LVLM capabilities via probability-based sampling to improve coherent visual search and multi-step reasoning.
ROVER introduces a learnable routing plugin for object-centric visual evidence in MLLMs via token triplets and differential attention, reporting gains on MM-GCoT and VideoEspresso when integrated into Qwen2.5-VL-7B.
Learns state-conditioned commitment depth in a 7B vision-language policy that jointly predicts actions and replan intervals, outperforming fixed-depth baselines and larger models on Sliding Puzzle and Sokoban while providing a theoretical dominance result.
ToolsRL trains MLLMs via a tool-specific then accuracy-focused RL curriculum to master visual tools for complex reasoning tasks.
Mull-Tokens are modality-agnostic latent tokens that enable free-form multimodal thinking and deliver up to 16% gains on spatial reasoning benchmarks.
Multimodal foundation models achieve respectable but sub-specialist performance on semantic vision tasks and weaker results on geometric tasks when evaluated through prompt chaining on established benchmarks.
ViGoRL introduces visually grounded RL that anchors reasoning steps to image coordinates and uses multi-turn zooming to outperform standard RL and supervised baselines on spatial and GUI reasoning benchmarks.
CoT-VLA is a 7B VLA that generates future visual frames autoregressively as planning goals before actions, outperforming prior VLAs by 17% on real-world tasks and 6% in simulation.
Iterative SFT-RL cycles enable a 7B LVLM to develop sophisticated visual chain-of-thought reasoning and improve performance on math and general reasoning benchmarks.
MVoT lets multimodal models create coherent images during chain-of-thought reasoning via a token discrepancy loss, yielding competitive or better results than text-only CoT on dynamic spatial tasks.
MathVis-Fine proposes a dataset with fine-grained visual annotations and dependency ratings plus a progressive two-stage training paradigm to align visual supervision with sample-specific necessity in multimodal mathematical reasoning.
ReFineVLA adds teacher-generated reasoning steps to VLA training and reports state-of-the-art success rates on SimplerEnv WidowX and Google Robot benchmarks.
MAG-3D is a training-free multi-agent framework that coordinates planning, grounding, and coding agents with off-the-shelf VLMs to achieve grounded 3D reasoning and state-of-the-art benchmark results.
A literature survey that introduces a taxonomy for LLM reasoning paradigms, analyzes methodological trends, and synthesizes failure modes from over 300 papers.
citing papers explorer
-
When to Re-Commit: Temporal Abstraction Discovery for Long-Horizon Vision-Language Reasoning
Learns state-conditioned commitment depth in a 7B vision-language policy that jointly predicts actions and replan intervals, outperforming fixed-depth baselines and larger models on Sliding Puzzle and Sokoban while providing a theoretical dominance result.