Introduces the TVR active viewpoint-matching task and TVRBench indoor simulation benchmark, where foundation models start at low single-digit success rates but reach 51.4% after visual-action SFT and multi-turn GRPO post-training.
super hub Mixed citations
GPT-4o System Card
Mixed citation behavior. Most common role is background (53%).
803
Pith papers citing it
Background
53% of classified citations
abstract
GPT-4o is an autoregressive omni model that accepts as input any combination of text, audio, image, and video, and generates any combination of text, audio, and image outputs. It's trained end-to-end across text, vision, and audio, meaning all inputs and outputs are processed by the same neural network. GPT-4o can respond to audio inputs in as little as 232 milliseconds, with an average of 320 milliseconds, which is similar to human response time in conversation. It matches GPT-4 Turbo performance on text in English and code, with significant improvement on text in non-English languages, while also being much faster and 50\% cheaper in the API. GPT-4o is especially better at vision and audio understanding compared to existing models. In line with our commitment to building AI safely and consistent with our voluntary commitments to the White House, we are sharing the GPT-4o System Card, which includes our Preparedness Framework evaluations. In this System Card, we provide a detailed look at GPT-4o's capabilities, limitations, and safety evaluations across multiple categories, focusing on speech-to-speech while also evaluating text and image capabilities, and measures we've implemented to ensure the model is safe and aligned. We also include third-party assessments on dangerous capabilities, as well as discussion of potential societal impacts of GPT-4o's text and vision capabilities.
hub tools
citation-role summary
background 97
baseline 51
method 23
dataset 3
citation-polarity summary
claims ledger
- abstract GPT-4o is an autoregressive omni model that accepts as input any combination of text, audio, image, and video, and generates any combination of text, audio, and image outputs. It's trained end-to-end across text, vision, and audio, meaning all inputs and outputs are processed by the same neural network. GPT-4o can respond to audio inputs in as little as 232 milliseconds, with an average of 320 milliseconds, which is similar to human response time in conversation. It matches GPT-4 Turbo performance on text in English and code, with significant improvement on text in non-English languages, while
authors
co-cited works
representative citing papers
VideoFDB is a new benchmark and LM-as-judge framework for evaluating full-duplex audio-visual-to-audio-visual conversational agents on nonverbal dynamics from real video calls.
M³Att poisons medical multimodal RAG by pairing covert textual misinformation with query-agnostic visual perturbations that increase retrieval of the bad content, causing LLMs to generate clinically plausible but incorrect responses.
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%.
CHASM is a new benchmark dataset showing that existing multimodal large language models fail to reliably detect covert advertisements on Chinese social media even after fine-tuning.
HalluAudio is the first large-scale benchmark spanning speech, environmental sound, and music that uses human-verified QA pairs, adversarial prompts, and mixed-audio tests to measure hallucinations in large audio-language models.
EVE enables verifiable self-evolution of MLLMs by using a Challenger-Solver architecture to generate dynamic executable visual transformations that produce VQA problems with absolute execution-verified ground truth.
Harmful skills in open agent ecosystems raise average harm scores from 0.27 to 0.76 across six LLMs by lowering refusal rates when tasks are presented via pre-installed skills.
ReConText3D is the first replay-memory framework for continual text-to-3D generation that prevents catastrophic forgetting on new textual categories while preserving quality on previously seen classes.
MMRareBench provides 1,756 QA pairs and 7,958 images from PMC rare-disease cases to evaluate 23 MLLMs, revealing low treatment-planning scores and medical models underperforming general models on multi-image tasks due to capacity dilution.
DialBGM is a new benchmark dataset revealing that existing AI models fall far short of human performance when recommending fitting background music for open-domain conversations.
EgoSound is a new benchmark with 7315 QA pairs across seven tasks to evaluate egocentric sound understanding in multimodal large language models.
VLRS-Bench is the first benchmark dedicated to complex vision-language reasoning in remote sensing, with 2000 QA pairs across 14 tasks in cognition, decision, and prediction dimensions.
SwissGov-RSD is the first naturalistic cross-lingual document-level benchmark with human token-level semantic difference annotations, on which both LLMs and encoders show a large performance gap relative to simpler settings.
CritPt benchmark shows state-of-the-art LLMs reach only 5.7% average accuracy on full-scale unpublished physics research tasks, rising to about 10% with coding tools.
Flow-GRPO is the first online RL method for flow matching models, raising GenEval accuracy from 63% to 95% and text-rendering accuracy from 59% to 92% with little reward hacking.
Molmo VLMs trained on newly collected PixMo open datasets achieve state-of-the-art performance among open-weight models and surpass multiple proprietary VLMs including Claude 3.5 Sonnet and Gemini 1.5 Pro.
LiveBench is a contamination-limited LLM benchmark with auto-scored challenging tasks from recent sources across math, coding, reasoning and more, where top models score below 70%.
OmniCoT is a new panoramic reasoning benchmark with 6.7K eval, 1K real, and 14.3K training examples plus a two-stage SFT+GRPO training method to enforce global 360-degree consistency.
MuseBench shows state-of-the-art MLLMs achieve only 48.29% accuracy on intent-level audiovisual arts understanding versus 87.18% for human experts.
A diagnostic framework called EPC reveals that proprietary LLM evaluators can exhibit large preference shifts between versions, as evidenced by a GPT-4o May-to-June drift that inverted study conclusions, rendering single-snapshot evaluations unreliable.
PhyEditBench is a new benchmark for physics-aware image editing with real and synthetic instances plus a training-free PhyWorld baseline that uses test-time scaling to outperform SOTA models.
CrypFormBench is a new benchmark jointly covering symbolic and computational security to evaluate LLMs on five formal analysis capabilities, with results showing top model Claude-3.5 scores 48.7/100 and most models struggling on generation, transformation, and correction.
SafeGen-Bench is a benchmark with 10 malicious categories that evaluates conditional T2V models on paired start frames and text prompts, finding unsafety scores up to 44.5 and 80% guardrail failure rate.
citing papers explorer
-
GeoBridge: A Semantic-Anchored Multi-View Foundation Model Bridging Images and Text for Geo-Localization
GeoBridge introduces a semantic-anchor mechanism using text to bridge multi-view image features for bidirectional cross-view and language-to-image geo-localization, supported by the new GeoLoc dataset of over 50,000 aligned pairs.
-
PhotoFramer: Multi-modal Image Composition Instruction
PhotoFramer is a multi-modal model that jointly produces textual composition instructions and illustrative corrected images from poorly framed inputs.
-
LongVT: Incentivizing "Thinking with Long Videos" via Native Tool Calling
LongVT adds native video-cropping tool calling to LMMs for interleaved multimodal chain-of-tool-thought reasoning on long videos and releases VideoSIAH data for training and evaluation.
-
Boosting Reasoning in Large Multimodal Models via Activation Replay
Activation Replay boosts multimodal reasoning in post-trained LMMs by replaying low-entropy activations from base models to RLVR counterparts at test time via visual token manipulation.
-
GraphMind: Theorem Selection and Conclusion Generation Framework with Dynamic GNN for LLM Reasoning
GraphMind models multi-step reasoning as an evolving heterogeneous graph, using GNN encoding and semantic matching to select theorems and generate conclusions iteratively, reporting performance gains over baselines on QA datasets.
-
MiMo-Embodied: X-Embodied Foundation Model Technical Report
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.
-
Look, Zoom, Understand: The Robotic Eyeball for Embodied Perception
EyeVLA transfers open-world VLM understanding to a PTZ camera control policy via hierarchical action tokens and GRPO reinforcement learning, reaching 96% task completion on 50 real scenes with only 500 training samples.
-
UniSER: A Foundation Model for Unified Soft Effects Removal
UniSER is a unified diffusion transformer foundation model that removes diverse soft image degradations by training on a large curated dataset of semi-transparent occlusions with fine-grained controls.
-
REVISOR: Beyond Textual Reflection, Towards Multimodal Introspective Reasoning in Long-Form Video Understanding
REVISOR adds multimodal visual-text reflection and a Dual Attribution Decoupled Reward to improve long-form video reasoning in MLLMs without extra supervised fine-tuning.
-
Evolve the Method, Not the Prompts: Evolutionary Synthesis of Jailbreak Attacks on LLMs
EvoSynth evolves code-based jailbreak algorithms via multi-agent self-correction, reaching 85.5% ASR on Claude-Sonnet-4.5 and 95.9% average across targets with greater diversity.
-
Stress Testing Factual Consistency Metrics for Long-Document Summarization
Short-form factual consistency metrics produce inconsistent scores on semantically equivalent long-document summaries and lose reliability on information-dense claims.
-
DeepEyesV2: Toward Agentic Multimodal Model
DeepEyesV2 uses a two-stage cold-start plus reinforcement learning pipeline to produce an agentic multimodal model that adaptively invokes tools and outperforms direct RL on real-world reasoning benchmarks.
-
Cambrian-S: Towards Spatial Supersensing in Video
Cambrian-S introduces VSI-SUPER benchmarks for long-horizon spatial recall and counting, shows data scaling yields 30% gains on existing tests, and demonstrates a self-supervised next-latent predictor using surprise outperforms baselines on the new spatial supersensing tasks.
-
Speculative Coupled Decoding for Training-Free Lossless Acceleration of Autoregressive Visual Generation
Speculative Coupled Decoding stabilizes draft sampling in Speculative Jacobi Decoding via an information-theoretic coupling step, delivering up to 4.2x image and 13.6x video speedups with no quality loss or training.
-
StableSketcher: Enhancing Diffusion Model for Pixel-based Sketch Generation via Visual Question Answering Feedback
StableSketcher improves text-to-sketch generation by fine-tuning a diffusion VAE and adding a VQA-based RL reward, while releasing the SketchDUO dataset of sketches with captions and QA pairs.
-
Structured and Abstractive Reasoning on Multi-modal Relational Knowledge Images
Authors build a synthetic data generator and two-stage training pipeline for structured abstractive reasoning on multi-modal relational knowledge images, releasing STAR-64K and showing 3B/7B models outperforming GPT-4o.
-
Can Old Tests Do New Tricks for Resolving SWE Issues?
TestPrune minimizes regression test suites to improve bug reproduction and patch validation in LLM-based agentic repair pipelines, delivering 6-13% relative gains on SWE-Bench benchmarks at low API cost.
-
AtlasKV: Augmenting LLMs with Billion-Scale Knowledge Graphs in 20GB VRAM
AtlasKV integrates billion-scale KGs into LLMs parametrically with sub-linear complexity and low memory by converting triples into key-value representations handled by the model's attention.
-
Mind-Paced Speaking: A Dual-Brain Approach to Real-Time Reasoning in Spoken Language Models
MPS proposes a dual-brain architecture separating formulation reasoning from articulation to achieve real-time CoT in SLMs with accuracy comparable to full pre-computation but much lower latency.
-
Test-Time Matching: Unlocking Compositional Reasoning in Multimodal Models
Introduces group matching score for better evaluation of compositional reasoning and Test-Time Matching (TTM) algorithm for unsupervised self-improvement in multimodal models, achieving SOTA gains including surpassing GPT-4.1 and estimated human performance.
-
ARM: Discovering Agentic Reasoning Modules for Generalizable Multi-Agent Systems
ARM evolves specialized reasoning modules from basic CoT via tree search to serve as reusable components in multi-agent systems that generalize across models and domains without per-task re-optimization.
-
Inferring Dynamic Physical Properties from Video Foundation Models
Video foundation models infer dynamic physical properties such as elasticity, viscosity, and friction from videos at levels close to classical oracles while outperforming current MLLMs with suitable prompting.
-
Multi-LLM Orchestration for High-Quality Code Generation: Exploiting Complementary Model Strengths
PerfOrch is a four-agent multi-LLM system that uses offline profiling to build language-and-category rankings for routing tasks, achieving 97.19% and 95.83% pass@1 on HumanEval-X and EffiBench-X with generalization across benchmarks.
-
Mitigating Visual Context Degradation in Large Multimodal Models: A Training-Free Decoupled Agentic Framework
DRP decouples reasoning from perception in LMMs by using an LLM reasoner to query an LMM observer for visual details as needed, reducing visual grounding loss.
-
ChatInject: Abusing Chat Templates for Prompt Injection in LLM Agents
ChatInject exploits LLM chat template structures to boost indirect prompt injection success rates on agents from ~5-15% to 32-52% across benchmarks, with multi-turn persuasion variants performing best.
-
StableToken: A Noise-Robust Semantic Speech Tokenizer for Resilient SpeechLLMs
StableToken introduces a multi-branch architecture with bit-wise voting to create noise-robust semantic speech tokens, achieving lower Unit Edit Distance and better SpeechLLM robustness than prior single-path tokenizers.
-
RISK: A Framework for GUI Agents in E-commerce Risk Management
RISK introduces a dataset, benchmark, and R1-style RL fine-tuning for GUI agents that achieve 6.8-8.8% offline gains and 70.5% online task success in e-commerce risk management using 7.2% of baseline parameters.
-
Benchmarking and Mitigating Sycophancy in Medical Vision Language Models
The paper benchmarks sycophancy in medical VLMs using hierarchical VQA templates and proposes VIPER to filter non-evidence social cues, reducing sycophancy while preserving interpretability.
-
Compute as Teacher: Turning Inference Compute Into Reference-Free Supervision
Parallel inference rollouts aggregated into pseudo-references enable reference-free RL supervision that matches expert-annotated performance on health tasks while using 9x less test-time compute.
-
PiERN: Token-Level Routing for Integrating High-Precision Computation and Reasoning
PiERN proposes token-level routing of physically-isolated experts to embed high-precision computation directly into LLMs, reporting higher accuracy and lower latency, token count, and energy use than fine-tuning or multi-agent baselines.
-
ScaleDoc: Scaling LLM-based Predicates over Large Document Collections
ScaleDoc achieves over 2x end-to-end speedup and up to 85% fewer LLM invocations for semantic predicates on large document collections via offline LLM representations, contrastive-trained proxy filtering, and adaptive cascades.
-
VideoAgent: Personalized Synthesis of Scientific Videos
VideoAgent is a modular framework that redefines scientific video synthesis as an intent-driven planning problem and introduces the SciVidEval benchmark for multimodal quality and pedagogical utility.
-
VeriOS: Query-Driven Proactive Human-Agent-GUI Interaction for Trustworthy OS Agents
VeriOS-Agent is an OS agent that proactively queries humans in untrustworthy scenarios via a query-driven framework and three-stage training, achieving 19.72% higher step-wise success rate over baselines while preserving normal performance.
-
Disentangling Interaction and Bias Effects in Opinion Dynamics of Large Language Models
A Bayesian framework disentangles topic, agreement, and anchoring biases from interaction effects in LLM multi-turn dialogues, revealing convergence to attractors that shift with fine-tuning.
-
Learning to Conceal Risk: Controllable Multi-turn Red Teaming for LLMs in the Financial Domain
CoRT achieves 95% average attack success rate on nine LLMs by using iterative risk-concealing prompts and a controller that scores concealment levels on a new 522-instruction financial risk benchmark.
-
A Functionality-Grounded Benchmark for Evaluating Web Agents in E-commerce Domains
The paper proposes Amazon-Bench, a functionality-grounded benchmark for web agents in e-commerce that generates diverse task queries from webpage elements and evaluates both task performance and safety risks.
-
GoViG: Goal-Conditioned Visual Navigation Instruction Generation via Multimodal Reasoning
GoViG decomposes goal-conditioned navigation instruction generation into visual state prediction and instruction synthesis using an autoregressive multimodal LLM with one-pass and interleaved reasoning, showing gains on a new R2R-Goal dataset.
-
Training-Free Multimodal Large Language Model Orchestration
LLM Orchestration integrates modality experts via an LLM controller, cross-modal memory, and interaction layer to enable multimodal input-output without gradient-based training.
-
TARS: MinMax Token-Adaptive Preference Strategy for Hallucination Reduction in MLLMs
TARS uses token-adaptive min-max preference optimization and FFT-based spectral regularization to cut hallucination rates in MLLMs from 26.4% to 13.2% with only 4.8k samples, outperforming standard DPO and larger data-augmented baselines.
-
UniDomain: Pretraining a Unified PDDL Domain from Real-World Demonstrations for Generalizable Robot Task Planning
UniDomain extracts atomic PDDL domains from 12,393 robot videos to create a unified domain of 3137 operators and 2875 predicates, then retrieves and fuses relevant parts to enable zero-shot planning on unseen real-world tasks.
-
PRISM: Programmatic Reasoning with Image Sequence Manipulation for LVLM Jailbreaking
PRISM decomposes harmful instructions into benign visual gadgets and directs LVLMs via prompts to compose them through reasoning into harmful outputs, achieving ASR over 0.90 on SafeBench.
-
ReGATE: Learning Faster and Better with Fewer Tokens in MLLMs
ReGATE introduces a teacher-student adaptive token elision method that reduces training tokens to 38% while matching or exceeding baseline accuracy on multimodal benchmarks.
-
Rubrics as Rewards: Reinforcement Learning Beyond Verifiable Domains
RaR uses aggregated rubric feedback as rewards in on-policy RL, delivering up to 31% relative gains on HealthBench and 7% on GPQA-Diamond versus direct Likert LLM-as-judge baselines.
-
Step-Audio 2 Technical Report
Step-Audio 2 integrates a latent audio encoder, reasoning-centric reinforcement learning, and discrete audio token generation into language modeling to deliver state-of-the-art performance on audio understanding and conversational benchmarks.
-
Vidar: Embodied Video Diffusion Model for Generalist Manipulation
Vidar shows that a video diffusion prior continuously pre-trained on 750K multi-view robot trajectories plus a label-free masked inverse dynamics adapter can generalize manipulation to new robot embodiments with 1% of typical demonstration data.
-
Mitigating Object Hallucinations via Sentence-Level Early Intervention
SENTINEL reduces MLLM object hallucinations by over 90% via sentence-level early intervention with detector-bootstrapped preference data and C-DPO loss, outperforming prior SOTA on hallucination and capability benchmarks.
-
GTA1: GUI Test-time Scaling Agent
GTA1 combines test-time scaling for action plan selection with RL-based grounding to achieve SOTA results on GUI agent benchmarks.
-
Does Math Reasoning Improve General LLM Capabilities? Understanding Transferability of LLM Reasoning
Math reasoning gains in LLMs rarely transfer to general domains; RL tuning generalizes while SFT causes forgetting and representation drift.
-
NaviAgent: Bilevel Planning on Tool Navigation Graph for Large-Scale Orchestration
NaviAgent decouples task planning from tool execution via a Tool World Navigation Model graph to improve scalability and success rates in LLM agents handling large tool ecosystems.
-
Mercury: Ultra-Fast Language Models Based on Diffusion
Mercury Coder diffusion LLMs achieve throughputs of 1109 and 737 tokens per second on H100 GPUs, up to 10x faster than frontier models with comparable quality.