S1-VL combines structured scientific reasoning with iterative image manipulation via code execution to reach state-of-the-art results on visual and scientific reasoning benchmarks.
hub Mixed citations
Intern-s1: A scientific multimodal foundation model
Mixed citation behavior. Most common role is background (67%).
27
Pith papers citing it
Background
67% of classified citations
hub tools
citation-role summary
background 4
baseline 1
method 1
citation-polarity summary
representative citing papers
V2X-QA provides a view-decoupled benchmark showing infrastructure views aid macroscopic traffic understanding while cooperative reasoning requires explicit cross-view alignment, with V2X-MoE as a routing-based baseline that improves performance.
ReCrit frames critic interaction as a correctness-transition problem and uses quadrant-based RL rewards to improve LLM performance on scientific reasoning benchmarks by rewarding corrections and robustness while penalizing sycophancy.
MolRecBench-Wild reveals that 18 existing OCSR models suffer severe performance drops on complex real-world academic molecular images compared with prior patent benchmarks.
DualComp uses a lightweight router to split visual token compression into a semantic stream with size-adaptive clustering and a geometric stream with path-tracing recovery, enabling low-cost high-fidelity UHR remote sensing interpretation.
PolyReal benchmark shows leading MLLMs perform well on polymer knowledge reasoning but drop sharply on practical tasks like lab safety analysis and raw data extraction.
Reinforcement learning after SFT conversion narrows the performance gap between sliding-window attention and full self-attention on math reasoning benchmarks while preserving linear complexity.
Sci-PRM is a tool-aware process reward model trained on the SCIPRM70K dataset to provide fine-grained supervision for scientific reasoning and shown to boost foundation models via Best-of-N selection and RL.
Stopping large reasoning models at the first correct reasoning prefix improves accuracy up to 21% by avoiding harmful overthinking that destabilizes correct trajectories.
MACReD is a multi-agent collaborative reasoning framework for reaction diagram parsing that reports state-of-the-art F1 scores of 75.2% and 84.6% on the RxnScribe benchmark.
Maestro uses outcome-based RL to train a lightweight policy that orchestrates ensembles of frozen expert models and skills, reporting 70.1% average accuracy across ten multimodal benchmarks and outperforming GPT-5 and Gemini-2.5-Pro while generalizing to unseen components.
LGBO integrates LLM semantic preferences continuously into Bayesian optimization iterations, with a theoretical worst-case guarantee and empirical gains including 90% of best value in 6 iterations on a wet-lab battery task.
AOT-POT adaptively reshapes complex PDE solution operators via input-dependent transformations and parallel stream mixing to enable effective large-scale pre-training, yielding SOTA results on 12 benchmarks with minimal added parameters.
Seirênes trains LLMs via adversarial self-play to generate and overcome evolving distractions, producing gains of 7-10 points on math reasoning benchmarks and exposing blind spots in larger models.
Meow-Omni 1 is a quad-modal MLLM that fuses video, audio, physiological time-series, and text to achieve 71.16% accuracy on feline intent recognition in the new MeowBench benchmark.
Degraded image resolution in MLLMs bypasses safety alignments via cognitive overload, raising jailbreak rates across perturbations.
SafeSci creates a large objective benchmark and training resource that reveals safety weaknesses in current LLMs for science and demonstrates measurable improvement through targeted fine-tuning.
FinReasoning is a hierarchical benchmark that decomposes LLM financial research capabilities into semantic consistency, data alignment, and deep insight, revealing model-type differences in auditing versus insight generation.
Thinking with Drafting reconceptualizes visual reasoning as optical decompression by forcing models to draft mental models into executable DSL code for deterministic self-verification on the VisAlg benchmark.
DeepSearch embeds MCTS into RLVR training with global frontier selection, entropy guidance, and adaptive replay to achieve 62.95% average accuracy on math reasoning benchmarks while using 5.7x fewer GPU hours than extended training.
RSICCLLM introduces a post-training framework with RSICI dataset, difference-aware supervised fine-tuning, and dual-negative preference optimization that claims to outperform much larger models on remote sensing image change captioning.
RigPI combines VLM initialization with two-stage gradient-based optimization in differentiable simulation to estimate dynamic parameters of rigid bodies from real robot interactions.
SciCore-Mol augments LLMs with three integrated modules for molecular perception, latent diffusion generation, and reaction reasoning, claiming an 8B open model competes with or exceeds proprietary systems on chemical tasks.
S1-Omni-Image unifies scientific image understanding, generation and editing via a think-before-generate paradigm on top of S1-VL-32B, trained on a 314K-sample SciGenEdit dataset, and reports SOTA results on multiple generation and editing benchmarks.
citing papers explorer
-
S1-VL: Scientific Multimodal Reasoning Model with Thinking-with-Images
S1-VL combines structured scientific reasoning with iterative image manipulation via code execution to reach state-of-the-art results on visual and scientific reasoning benchmarks.
-
V2X-QA: A Comprehensive Reasoning Dataset and Benchmark for Multimodal Large Language Models in Autonomous Driving Across Ego, Infrastructure, and Cooperative Views
V2X-QA provides a view-decoupled benchmark showing infrastructure views aid macroscopic traffic understanding while cooperative reasoning requires explicit cross-view alignment, with V2X-MoE as a routing-based baseline that improves performance.
-
ReCrit: Transition-Aware Reinforcement Learning for Scientific Critic Reasoning
ReCrit frames critic interaction as a correctness-transition problem and uses quadrant-based RL rewards to improve LLM performance on scientific reasoning benchmarks by rewarding corrections and robustness while penalizing sycophancy.
-
MolRecBench-Wild: A Real-World Benchmark for Optical Chemical Structure Recognition
MolRecBench-Wild reveals that 18 existing OCSR models suffer severe performance drops on complex real-world academic molecular images compared with prior patent benchmarks.
-
Semantic-Geometric Dual Compression: Training-Free Visual Token Reduction for Ultra-High-Resolution Remote Sensing Understanding
DualComp uses a lightweight router to split visual token compression into a semantic stream with size-adaptive clustering and a geometric stream with path-tracing recovery, enabling low-cost high-fidelity UHR remote sensing interpretation.
-
PolyReal: A Benchmark for Real-World Polymer Science Workflows
PolyReal benchmark shows leading MLLMs perform well on polymer knowledge reasoning but drop sharply on practical tasks like lab safety analysis and raw data extraction.
-
Architecture-Aware Reinforcement Learning Makes Sliding-Window Attention Competitive in Math Reasoning
Reinforcement learning after SFT conversion narrows the performance gap between sliding-window attention and full self-attention on math reasoning benchmarks while preserving linear complexity.
-
SCI-PRM: A Tool Aware Process Reward Model for Scientific Reasoning Verification
Sci-PRM is a tool-aware process reward model trained on the SCIPRM70K dataset to provide fine-grained supervision for scientific reasoning and shown to boost foundation models via Best-of-N selection and RL.
-
Thinking Past the Answer: Evaluating Harmful Overthinking in Large Reasoning Models
Stopping large reasoning models at the first correct reasoning prefix improves accuracy up to 21% by avoiding harmful overthinking that destabilizes correct trajectories.
-
MACReD: A Multi-Agent Collaborative Reasoning Framework for Reaction Diagram Parsing
MACReD is a multi-agent collaborative reasoning framework for reaction diagram parsing that reports state-of-the-art F1 scores of 75.2% and 84.6% on the RxnScribe benchmark.
-
Maestro: Reinforcement Learning to Orchestrate Hierarchical Model-Skill Ensembles
Maestro uses outcome-based RL to train a lightweight policy that orchestrates ensembles of frozen expert models and skills, reporting 70.1% average accuracy across ten multimodal benchmarks and outperforming GPT-5 and Gemini-2.5-Pro while generalizing to unseen components.
-
Unleashing LLMs in Bayesian Optimization: Preference-Guided Framework for Scientific Discovery
LGBO integrates LLM semantic preferences continuously into Bayesian optimization iterations, with a theoretical worst-case guarantee and empirical gains including 90% of best value in 6 iterations on a wet-lab battery task.
-
AOT-POT: Adaptive Operator Transformation for Large-Scale PDE Pre-training
AOT-POT adaptively reshapes complex PDE solution operators via input-dependent transformations and parallel stream mixing to enable effective large-scale pre-training, yielding SOTA results on 12 benchmarks with minimal added parameters.
-
Seir\^enes: Adversarial Self-Play with Evolving Distractions for LLM Reasoning
Seirênes trains LLMs via adversarial self-play to generate and overcome evolving distractions, producing gains of 7-10 points on math reasoning benchmarks and exposing blind spots in larger models.
-
Meow-Omni 1: A Multimodal Large Language Model for Feline Ethology
Meow-Omni 1 is a quad-modal MLLM that fuses video, audio, physiological time-series, and text to achieve 71.16% accuracy on feline intent recognition in the new MeowBench benchmark.
-
Hard to Read, Easy to Jailbreak: How Visual Degradation Bypasses MLLM Safety Alignment
Degraded image resolution in MLLMs bypasses safety alignments via cognitive overload, raising jailbreak rates across perturbations.
-
SafeSci: Safety Evaluation of Large Language Models in Science Domains and Beyond
SafeSci creates a large objective benchmark and training resource that reveals safety weaknesses in current LLMs for science and demonstrates measurable improvement through targeted fine-tuning.
-
FinReasoning: A Hierarchical Benchmark for Reliable Financial Research Reporting
FinReasoning is a hierarchical benchmark that decomposes LLM financial research capabilities into semantic consistency, data alignment, and deep insight, revealing model-type differences in auditing versus insight generation.
-
Thinking with Drafting: Optical Decompression via Logical Reconstruction
Thinking with Drafting reconceptualizes visual reasoning as optical decompression by forcing models to draft mental models into executable DSL code for deterministic self-verification on the VisAlg benchmark.
-
DeepSearch: Overcome the Bottleneck of Reinforcement Learning with Verifiable Rewards via Monte Carlo Tree Search
DeepSearch embeds MCTS into RLVR training with global frontier selection, entropy guidance, and adaptive replay to achieve 62.95% average accuracy on math reasoning benchmarks while using 5.7x fewer GPU hours than extended training.
-
RSICCLLM: A Multimodal Large Language Model for Remote Sensing Image Change Captioning
RSICCLLM introduces a post-training framework with RSICI dataset, difference-aware supervised fine-tuning, and dual-negative preference optimization that claims to outperform much larger models on remote sensing image change captioning.
-
RigPI: Dynamic Parameter Identification of Rigid Body via VLM-Seeded Differentiable Simulation
RigPI combines VLM initialization with two-stage gradient-based optimization in differentiable simulation to estimate dynamic parameters of rigid bodies from real robot interactions.
-
SciCore-Mol: Augmenting Large Language Models with Pluggable Molecular Cognition Modules
SciCore-Mol augments LLMs with three integrated modules for molecular perception, latent diffusion generation, and reaction reasoning, claiming an 8B open model competes with or exceeds proprietary systems on chemical tasks.
-
S1-Omni-Image: A Unified Model for Scientific Image Understanding, Generation, and Editing
S1-Omni-Image unifies scientific image understanding, generation and editing via a think-before-generate paradigm on top of S1-VL-32B, trained on a 314K-sample SciGenEdit dataset, and reports SOTA results on multiple generation and editing benchmarks.
-
A Survey of Reinforcement Learning for Large Reasoning Models
A survey compiling RL methods, challenges, data resources, and applications for enhancing reasoning in large language models and large reasoning models since DeepSeek-R1.
-
Earth Science Foundation Models: From Perception to Reasoning and Discovery
A review of Earth science foundation models covering capability evolution from perception to discovery, applications across atmosphere/hydrosphere/lithosphere/biosphere/anthroposphere/cryosphere, over 200 datasets, and key challenges.
- EgoMind: Activating Spatial Cognition through Linguistic Reasoning in MLLMs