FigSIM is the first annotated dataset for fine-grained suicide severity and figurative language in suicide memes, accompanied by benchmarks on 16 unimodal and multimodal models.
super hub Mixed citations
InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency
Mixed citation behavior. Most common role is background (50%).
434
Pith papers citing it
Background
50% of classified citations
abstract
We introduce InternVL 3.5, a new family of open-source multimodal models that significantly advances versatility, reasoning capability, and inference efficiency along the InternVL series. A key innovation is the Cascade Reinforcement Learning (Cascade RL) framework, which enhances reasoning through a two-stage process: offline RL for stable convergence and online RL for refined alignment. This coarse-to-fine training strategy leads to substantial improvements on downstream reasoning tasks, e.g., MMMU and MathVista. To optimize efficiency, we propose a Visual Resolution Router (ViR) that dynamically adjusts the resolution of visual tokens without compromising performance. Coupled with ViR, our Decoupled Vision-Language Deployment (DvD) strategy separates the vision encoder and language model across different GPUs, effectively balancing computational load. These contributions collectively enable InternVL3.5 to achieve up to a +16.0\% gain in overall reasoning performance and a 4.05$\times$ inference speedup compared to its predecessor, i.e., InternVL3. In addition, InternVL3.5 supports novel capabilities such as GUI interaction and embodied agency. Notably, our largest model, i.e., InternVL3.5-241B-A28B, attains state-of-the-art results among open-source MLLMs across general multimodal, reasoning, text, and agentic tasks -- narrowing the performance gap with leading commercial models like GPT-5. All models and code are publicly released.
hub tools
citation-role summary
background 54
baseline 41
method 12
dataset 1
other 1
citation-polarity summary
claims ledger
- abstract We introduce InternVL 3.5, a new family of open-source multimodal models that significantly advances versatility, reasoning capability, and inference efficiency along the InternVL series. A key innovation is the Cascade Reinforcement Learning (Cascade RL) framework, which enhances reasoning through a two-stage process: offline RL for stable convergence and online RL for refined alignment. This coarse-to-fine training strategy leads to substantial improvements on downstream reasoning tasks, e.g., MMMU and MathVista. To optimize efficiency, we propose a Visual Resolution Router (ViR) that dynami
authors
co-cited works
representative citing papers
EPIC-Bench is a new fine-grained benchmark that shows leading VLMs struggle with multi-target counting, part-whole relations, and affordance detection in real-world embodied visual grounding tasks.
MI-CXR is a new benchmark that shows state-of-the-art vision-language models achieve only 29.3% accuracy on longitudinal reasoning tasks across multi-visit chest X-ray sequences.
LongAct benchmark evaluates long-horizon household task execution from free-form instructions; HoloMind agent raises performance but top VLMs still reach only 59% goal completion and 16% full-task success.
SenseBench is the first physics-based benchmark with 10K+ instances and dual protocols to evaluate VLMs on remote sensing low-level perception and diagnostic description, revealing domain bias and specific failure modes.
EgoMemReason is a new benchmark showing that even the best multimodal models achieve only 39.6% accuracy on reasoning tasks that require integrating sparse evidence across days in egocentric video.
RuleSafe-VL creates 2,166 rule-conditioned cases from 93 atomic rules and 92 relations across three policy families to diagnose where VLMs fail at rule-based content moderation reasoning.
TraceAV-Bench is the first benchmark for multi-hop trajectory reasoning over long audio-visual videos, showing top models reach only 51-68% accuracy with substantial room for improvement.
MedHorizon benchmark reveals current multimodal LLMs achieve only 41.1% accuracy on long medical videos due to failures in sparse evidence retrieval and procedural reasoning.
Creates the first benchmark dataset integrating papers, slides, videos, and presentations for evaluating AI models on fine-grained multimodal correspondences in science.
MLLMs exhibit a Mirage effect by bypassing circuit diagrams in favor of header semantics for Verilog generation; VeriGround with identifier anonymization and D-ORPO training reaches 46% Functional Pass@1 while refusing blank images at >92%.
MM-JudgeBench shows substantial cross-lingual performance variance in 22 LVLM judges, with model size and architecture as poor predictors of multilingual robustness.
VLMs hallucinate by prioritizing contradictory on-screen text over visual content, addressed via the VisualTextTrap benchmark with 6,057 human-validated samples and the VTHM-MoE dual-encoder framework using dimension-specific experts and adaptive routing.
RefereeBench shows that even the strongest video MLLMs reach only around 60% accuracy on multi-sport refereeing tasks and struggle with rule application and temporal grounding.
Dental-TriageBench is the first expert-annotated multimodal benchmark for hierarchical dental triage and shows a substantial performance gap between 19 MLLMs and junior dentists, especially on multi-domain referral cases.
VAREX benchmark shows structured output compliance limits models under 4B parameters more than extraction ability, with layout-preserving text giving the largest accuracy gains over images.
Molmo2 delivers state-of-the-art open-weight video VLMs with new grounding datasets and training methods that outperform prior open models and match or exceed some proprietary ones on pointing and tracking tasks.
ToG-Bench is the first benchmark for task-oriented spatio-temporal video grounding in egocentric videos, with explicit-implicit dual grounding and one-to-many object scenarios across 100 ScanNet clips and 2704 instructions.
AnyGroundBench is a domain-adaptation benchmark for spatio-temporal video grounding across animal, industry, sports, surgery, and public security domains that finds 15 state-of-the-art VLMs fail in zero-shot and ICL settings.
LongEgoRefer is a new benchmark of 1,498 referring expressions in 45-minute average egocentric videos that exposes the failure of existing Video REC models on sparse long-form spatio-temporal grounding.
A large examination-level ultrasound dataset with long-form reports enables simple LVLM fine-tuning to outperform prior complex methods.
Introduces Anom-π framework for active video anomaly understanding via interleaved policy optimization and iDPO under weak supervision, claiming a 2B model outperforms larger SOTA VAU models.
EgoGapBench shows humans reliably select egocentric actions in multi-agent scenes while MLLMs systematically choose other agents' actions, and standard egocentric training data fails to close the gap.
EgoSafetyBench shows VLMs reliably spot hazard-containing videos but miss specific contextual hazards and are degraded by misleading in-scene text.
citing papers explorer
-
When Robots Do the Chores: A Benchmark and Agent for Long-Horizon Household Task Execution
LongAct benchmark evaluates long-horizon household task execution from free-form instructions; HoloMind agent raises performance but top VLMs still reach only 59% goal completion and 16% full-task success.
-
RuleSafe-VL: Evaluating Rule-Conditioned Decision Reasoning in Vision-Language Content Moderation
RuleSafe-VL creates 2,166 rule-conditioned cases from 93 atomic rules and 92 relations across three policy families to diagnose where VLMs fail at rule-based content moderation reasoning.
-
From Senses to Decisions: The Information Flow of Auditory and Visual Perception in Multimodal LLMs
AVLLMs route audio-visual information sequentially in video tasks and via parallel streams for interleaved items, allowing early token discard with little performance loss across models and scales.
-
Measuring Cross-Modal Synergy: A Benchmark for VLM Explainability
Introduces Synergistic Faithfulness metric based on Shapley Interaction Index to evaluate cross-modal synergy in VLM explainers, revealing over-reliance on visual salience in existing methods.
-
Done, But Not Sure: Disentangling World Completion from Self-Termination in Embodied Agents
VIGIL decouples world-state completion from terminal commitment in embodied agents, exposing up to 19.7 pp gaps in benchmark success despite comparable execution across 20 models.
-
CrossCult-KIBench: A Benchmark for Cross-Cultural Knowledge Insertion in MLLMs
CrossCult-KIBench is a new benchmark for evaluating cross-cultural knowledge insertion in MLLMs, paired with the MCKI baseline method, showing current approaches fail to balance adaptation and preservation.
-
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.
-
MirrorBench: Evaluating Self-centric Intelligence in MLLMs by Introducing a Mirror
MirrorBench reveals that leading MLLMs perform far below humans on tasks requiring self-referential perception and representation, even at the simplest level.
-
RiskWebWorld: A Realistic Interactive Benchmark for GUI Agents in E-commerce Risk Management
RiskWebWorld is the first realistic interactive benchmark for GUI agents in e-commerce risk management, revealing a large gap between generalist and specialized models plus RL gains.
-
TAVR-VLM: Risk-Conditioned Causal Grounding for Hallucination-Resistant Report Generation
TAVR-VLM introduces Risk-Conditioned Causal Grounding Attention to achieve SOTA AUROC 0.896, CIDEr 0.936, and 8.1% hallucination rate on a 1,482-patient TAVR cohort.
-
RoboPIN: Grounded Embodied Reasoning via Pinned Chain-of-Thought
Introduces PinCoT paradigm with visual reasoning anchors, builds PIN-170K dataset via automated pipeline, and trains 4B RoboPIN model via three-stage post-training to outperform 7B baselines by 12% on embodied reasoning benchmarks.
-
Reasoning for Mobile User Experience with Multimodal LLMs: Task, Benchmark, and Approach
Introduces UXBench benchmark for MLLM UI UX reasoning and UI-UX model achieving 0.7963 accuracy via RL enhancements on Qwen3-VL base.
-
Benchmark Everything Everywhere All at Once
Benchmark Agent is an autonomous agentic system that constructs benchmarks for LLMs and MLLMs via query analysis, subtask design, annotation and quality control, yielding 15 benchmarks with minimal human input.
-
Seeing Time: Benchmarking Chronological Reasoning and Shortcut Biases in Vision-Language Models
Introduces ChronoVision benchmark with three datasets showing VLMs rely on superficial cues such as color filters rather than genuine chronological reasoning.
-
Brick-Composer: Using MLLMs for Assembly with Diverse Bricks
Brick-Composer trains MLLMs on brick assembly via three signals, raising step-level success from under 1% to around 15% on the new BC-Bench benchmark.
-
Token Predictors Are Not Planners: Building Physically Grounded Causal Reasoners
Introduces a new diagnostic benchmark and million-scale reasoning corpus showing that training on explicit causal traces improves next-state prediction in embodied planning, with reported gains from data scaling.
-
MindClaw: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
MindClaw extends Theory of Mind to a closed-loop embodied setting with an optimized trigger skill that enables precise, timely intervention while avoiding unnecessary actions.
-
SPACENUM: Revisiting Spatial Numerical Understanding in VLMs
VLMs fail to ground numerical values in spatial perception on new bidirectional tasks, relying on shallow cues instead of coordinate-aware representations.
-
MPDocBench-Parse: Benchmarking Practical Multi-page Document Parsing
MPDocBench-Parse provides 433 annotated multi-page documents and an evaluation protocol covering text/table/formula extraction, merging, figure extraction, reading order, and heading hierarchy for realistic document parsing.
-
Chart-RL: Policy Optimization Reinforcement Learning for Enhanced Visual Reasoning in Chart Question Answering with Vision Language Models
Chart-RL uses RL policy optimization and LoRA to boost VLM chart reasoning, enabling a 4B model to reach 0.634 accuracy versus 0.580 for an 8B model with lower latency.
-
Cognitive Mismatch in Multimodal Large Language Models for Discrete Symbol Understanding
MLLMs exhibit a consistent recognition-reasoning inversion on discrete visual symbols across domains, underperforming on elementary perception while appearing competent on higher-level reasoning via linguistic compensation.
-
Revis: Sparse Latent Steering to Mitigate Object Hallucination in Large Vision-Language Models
REVIS reduces object hallucination in large vision-language models by about 19% via sparse orthogonal projection in latent space at suppression depths while keeping reasoning intact.
-
Harnessing Textual Refusal Directions for Multimodal Safety
Textual refusal directions generalize across modalities in MLLMs, enabling the training-free MARS method that corrects misalignment and improves safety while preserving utility.
-
Kairos: A Native World Model Stack for Physical AI
Kairos is a native world model stack using cross-embodiment pretraining, hybrid linear temporal attention with theoretical error bounds, and deployment-aware co-design, reporting top performance on embodied benchmarks.
-
AlloSpatial: Agentic Harness Framework for Spatial Reasoning in Foundation Models
AlloSpatial adds structured allocentric priors and a harness for tool-use and arbitration to improve spatial reasoning in foundation models, with 5-18% gains on VSI-Bench and MindCube in training-free settings and further gains after RL internalization.
-
Mind the Tool Failures: Achieving Synergistic Tool Gains for Medical Agents
A GRPO-based RL framework with probabilistic risk minimization, disagreement-aware synergy rewards, and entropy-guided sampling enables instance-level tool selection that closes the single-oracle risk gap on medical benchmarks.
-
DT2IT-MRM: Debiased Preference Construction and Iterative Training for Multimodal Reward Modeling
DT2IT-MRM proposes a debiased preference construction pipeline, T2I data reformulation, and iterative training to curate multimodal preference data, achieving SOTA on VL-RewardBench, Multimodal RewardBench, and MM-RLHF-RewardBench.
-
DREAM-R: Multimodal Speculative Reasoning with RL-Based Refined Drafting, Precise Verification, and Fully Parallel Execution
DREAM-R introduces RL-based draft alignment, ratio-based verification, and parallel execution to accelerate speculative reasoning in multimodal models while preserving accuracy.