Vision-language models show some human-like patterns in visual search effort (flat for features, rising for conjunctions) but diverge on target-present vs absent slopes and enumeration accuracy when reasoning tokens proxy reaction time.
hub
Vision language models are blind
20 Pith papers cite this work. Polarity classification is still indexing.
20
Pith papers citing it
hub tools
citation-role summary
background 1
dataset 1
method 1
citation-polarity summary
verdicts
UNVERDICTED 20representative citing papers
PuMVR benchmark shows VLMs exhibit script-dependent bias on Punjabi tasks with accuracy gaps up to 16% and script consistency rates as low as 24.8%, even when visual input is provided.
ZPPO improves distillation to small vision-language models by using binary and negative candidate prompts plus a replay buffer for hard questions, outperforming standard distillation and GRPO on a 31-benchmark suite with largest gains at the 0.8B scale.
SpatialAct benchmark shows VLMs handle isolated spatial reasoning but fail to maintain coherent spatial beliefs and produce reliable actions in multi-turn 3D interactions, underperforming humans.
Task conditioning suppresses safety-critical signal reporting in language and vision models that unconstrained versions report at higher rates, creating an inattentional gap that decouples benchmark safety from real-world safety.
FineSightBench reveals VLMs perceive patterns down to 12px but show persistent failures in fine-scale reasoning such as numeracy and sequencing.
MOSS-Video-Preview introduces a cross-attention architecture and synthesized real-time QA data to enable continuous perception, answer revision, and faster inference in video-language models compared to decoder-only designs.
Decomposes VLM distillation loss into orthogonal language and visual components and introduces Visual Gradient Steering to prioritize visual grounding over standard monolithic optimization.
Training VLMs to point via text induces serial processing that eliminates binding errors and enables compositional generalization on multi-object tasks.
VLMs achieve 53-97% on rearrangement planning but only 6-45% on occlusion and under 7% on reflections, with failures localized to visual token compression after the vision encoder.
Stronger VLM agents use mirror reflections for self-identification in controlled 3D tests, while weaker ones inspect but fail to extract or correctly attribute self-relevant information.
S2H-DPO generates hierarchical prompt-driven preference pairs to improve multi-image reasoning in VLMs while keeping single-image performance intact.
ReflectCAP distills model-specific hallucination and oversight patterns into Structured Reflection Notes that steer LVLMs toward more factual and complete image captions, reaching the Pareto frontier on factuality-coverage trade-offs.
MiMo-Embodied is a single foundation model that achieves state-of-the-art results on 17 embodied AI benchmarks and 12 autonomous driving benchmarks through multi-stage learning, curated data, and CoT/RL fine-tuning that produces positive cross-domain transfer.
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.
MathFlow decouples perception and inference stages in MLLMs for visual math, with a dedicated perception model delivering gains on the FlowVerse benchmark when paired with existing reasoners.
Omni is a multimodal model whose native training on diverse data types enables context unrolling, allowing explicit reasoning across modalities to better approximate shared knowledge and improve downstream performance.
IRA is a stochastic attention mechanism that regulates visual information injection in VLMs to yield smoother embedding trajectories and reduced attention sinks.
Seed1.5-VL is a compact multimodal model that sets new records on dozens of vision-language benchmarks and outperforms prior systems on agent-style tasks.
citing papers explorer
-
Seed1.5-VL Technical Report
Seed1.5-VL is a compact multimodal model that sets new records on dozens of vision-language benchmarks and outperforms prior systems on agent-style tasks.