UniQL is a human-verified benchmark providing aligned natural language questions and dialect-specific SQL queries for 16 SQL systems to evaluate cross-dialect generalization.
mega hub Mixed citations
GPT-4o System Card
Mixed citation behavior. Most common role is background (53%).
1025
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 98
baseline 52
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
mega hub controls
export citing contexts JSON
export graph JSON
export full bundle JSON
open full Pith review
annotated reader queued
Recognition alignment
counterfactual ablation
co-cited works
representative citing papers
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.
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%.
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.
Introduces a cost-aware paired protocol with six outcome groups and applies it to Dynamic-SAGE versus SAGE, reporting 7.5-point accuracy gain, 28% fewer tool calls, but 34% higher token use.
P2R decouples perception from reasoning in VLMs via a two-stage process and PRA-GRPO alternating RL training, reporting gains such as 93.2% on V-Star for the 4B model over its Qwen3-VL backbone.
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.
Identifies Screen Perception and Misused Channel attack surfaces in VLM-powered mobile agents and demonstrates seven attacks enabling arbitrary command execution on five frameworks without privileges.
citing papers explorer
-
UniQL: Towards Dialect-Universal Benchmarking for Text-to-SQL
UniQL is a human-verified benchmark providing aligned natural language questions and dialect-specific SQL queries for 16 SQL systems to evaluate cross-dialect generalization.
-
DialBGM: A Benchmark for Background Music Recommendation from Everyday Multi-Turn Dialogues
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.
-
Probing the Critical Point (CritPt) of AI Reasoning: a Frontier Physics Research Benchmark
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.
-
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.
-
ClinicalMC: A Benchmark for Multi-Course Clinical Decision-Making with Large Language Models
ClinicalMC is a benchmark of 1,275 Chinese and 5,804 English multi-course clinical samples across four stages, evaluated via a multi-agent framework on closed-source, open-source, and medical LLMs in static and dynamic settings.
-
AUDITFLOW: Executable Symbolic Environments for Structured Financial Reporting Verification
AuditFlow combines a graph-grounded symbolic environment with a multi-agent LLM setup to reach 82.09% joint audit accuracy on structured financial reports, 14.93 points above the strongest baseline.
-
OctoT2I: A Self-Evolving Agentic Text-to-Image Router
OctoT2I uses a no-supervision PSEL loop to discover model capability frontiers and route T2I tasks, reaching 0.96 GenEval score with 90.3% speedup over Flow-GRPO.
-
OmniMatBench: A Human-Calibrated Multimodal Reasoning Benchmark Across 19 Materials Science Subfields
OmniMatBench is a new human-calibrated benchmark for multimodal materials-science reasoning that reveals the best evaluated MLLM scores only 0.372 overall.
-
VeriTrip: A Verifiable Benchmark for Travel Planning Agents over Unstructured Web Corpora
VeriTrip is a new benchmark using a Multimodal Retrieval Base and Verifiable Knowledge Base to evaluate evidence-grounded reasoning and factual reliability in travel planning agents over unstructured multimodal web data.
-
Forecasting Scientific Progress with Artificial Intelligence
Introduces the CUSP benchmark across 4760 events and finds frontier AI models can pick plausible directions but fail to predict whether or when scientific advances will occur, with performance varying by domain and insensitive to training cutoffs.
-
Perception or Prejudice: Can MLLMs Go Beyond First Impressions of Personality?
Introduces the Grounded Personality Reasoning task and MM-OCEAN dataset to show that MLLMs frequently produce correct Big Five personality ratings without grounding them in observable video evidence.
-
Robotics-Inspired Guardrails for Foundation Models in Socially Sensitive Domains
Introduces the Grounded Observer framework that applies robotics-inspired formal constructs for runtime constraint enforcement on foundation model interaction trajectories in socially sensitive domains.
-
Hierarchical Attacks for Multi-Modal Multi-Agent Reasoning
HAM³ achieves up to 78.3% attack success rate on the GQA benchmark by hierarchically attacking perception, communication, and reasoning layers in multi-modal multi-agent systems.
-
PaperFit: Vision-in-the-Loop Typesetting Optimization for Scientific Documents
PaperFit uses rendered page images in a closed loop to diagnose and repair typesetting defects in LaTeX documents, outperforming baselines on a new benchmark of 200 papers.
-
EpiGraph: Building Generalists for Evidence-Intensive Epilepsy Reasoning in the Wild
EpiGraph creates a heterogeneous epilepsy knowledge graph that boosts LLM performance on clinical reasoning tasks by 30-41% in pharmacogenomics when used with Graph-RAG.
-
AHD Agent: Agentic Reinforcement Learning for Automatic Heuristic Design
AHD Agent trains a 4B-parameter LLM via agentic RL to actively use tools for automatic heuristic design, matching or exceeding larger baselines across eight domains with fewer evaluations.
-
LLM-guided Semi-Supervised Approaches for Social Media Crisis Data Classification
LG-CoTrain, an LLM-guided co-training method, outperforms classical semi-supervised baselines for crisis tweet classification in low-resource settings with 5-25 labeled examples per class.
-
Weblica: Scalable and Reproducible Training Environments for Visual Web Agents
Weblica scales RL training for visual web agents by building thousands of reproducible environments through HTTP caching for stable replays and LLM synthesis from real sites, yielding an 8B model that beats similar open baselines on navigation benchmarks.
-
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.
-
DataDignity: Training Data Attribution for Large Language Models
ScoringModel raises mean Recall@10 to 52.2 on the FakeWiki provenance benchmark from 35.0 for the best baseline, winning 41 of 45 model-by-condition comparisons and gaining 15.7 points on jailbreak-style queries.
-
Agentic-imodels: Evolving agentic interpretability tools via autoresearch
Agentic-imodels evolves scikit-learn regressors via an autoresearch loop to jointly boost predictive performance and LLM-simulatability, improving downstream agentic data science tasks by up to 73% on the BLADE benchmark.
-
Where Paths Split: Localized, Calibrated Control of Moral Reasoning in Large Language Models
A technique identifies minimal convergence-divergence points in LLM transformer blocks and calibrates residual-stream directions to achieve targeted ethical-framework control at inference time.
-
Enhancing Agent Safety Judgment: Controlled Benchmark Rewriting and Analogical Reasoning for Deceptive Out-of-Distribution Scenarios
ROME generates deceptive safety benchmarks that degrade LLM agent judgment performance, while ARISE uses analogical retrieval to improve safety decisions at inference time without retraining.
-
Investigating More Explainable and Partition-Free Compositionality Estimation for LLMs: A Rule-Generation Perspective
A rule-generation perspective lets LLMs write programs as rules for data mapping and applies complexity theory to estimate their compositionality, tested on string-to-grid tasks.
-
SciEval: A Benchmark for Automatic Evaluation of K-12 Science Instructional Materials
SciEval is a new benchmark of expert-annotated K-12 science lessons for LLM-based automatic evaluation, where zero-shot models perform poorly but fine-tuning yields up to 11% gains.
-
LLMs can persuade only psychologically susceptible humans on societal issues, via trust in AI and emotional appeals, amid logical fallacies
LLMs persuade only psychologically susceptible humans on societal issues through trust in AI and emotional appeals, while both sides rely on logical fallacies in roughly one out of every six conversational turns.
-
ADAPT: Benchmarking Commonsense Planning under Unspecified Affordance Constraints
ADAPT augments planners with affordance reasoning to raise task success in environments with unspecified and time-varying object affordances, and a LoRA-finetuned VLM backend beats GPT-4o on the new DynAfford benchmark.
-
GeoAgentBench: A Dynamic Execution Benchmark for Tool-Augmented Agents in Spatial Analysis
GeoAgentBench supplies a live execution environment and Plan-and-React architecture that lets tool-using AI agents handle multi-step GIS tasks more robustly than prior static evaluation methods.
-
TimeSeriesExamAgent: Creating Time Series Reasoning Benchmarks at Scale
TimeSeriesExamAgent combines templates and LLM agents to generate scalable time series reasoning benchmarks, demonstrating that current LLMs have limited performance on both abstract and domain-specific tasks.
-
PilotBench: A Benchmark for General Aviation Agents with Safety Constraints
PilotBench reveals that LLMs follow safety instructions well in flight trajectory prediction but deliver lower numerical precision than traditional forecasters, exposing a precision-controllability tradeoff.
-
How Independent are Large Language Models? A Statistical Framework for Auditing Behavioral Entanglement and Reweighting Verifier Ensembles
A new auditing framework reveals widespread behavioral entanglement among LLMs and shows that reweighting ensembles based on measured independence improves verification accuracy by up to 4.5%.
-
Pressure, What Pressure? Sycophancy Disentanglement in Language Models via Reward Decomposition
A five-term decomposed reward in GRPO training reduces sycophancy across models and generalizes to unseen pressure types by targeting pressure resistance and evidence responsiveness separately.
-
PERMA: Benchmarking Personalized Memory Agents via Event-Driven Preference and Realistic Task Environments
PERMA is a new benchmark using temporally ordered events, text variability, and linguistic alignment to evaluate LLM memory agents on persona consistency beyond simple retrieval.
-
ProactiveMobile: A Comprehensive Benchmark for Boosting Proactive Intelligence on Mobile Devices
ProactiveMobile is a new benchmark for proactive mobile agents that tests latent intent inference from context and executable API generation, where a fine-tuned 7B model reaches 19.15% success versus 15.71% for o1 and 7.39% for GPT-5.
-
Conv-FinRe: A Conversational and Longitudinal Benchmark for Utility-Grounded Financial Recommendation
Conv-FinRe is a new benchmark built from real market data and human trajectories that tests LLMs on generating utility-grounded stock rankings over fixed horizons while distinguishing rational analysis from behavioral mimicry or momentum.
-
SpatialBench: Benchmarking Multimodal Large Language Models for Spatial Cognition
SpatialBench creates a five-level framework and 15-task benchmark to measure hierarchical spatial reasoning in MLLMs, finding strong basic perception but weak symbolic reasoning, causal inference, and planning.
-
ToolPRM: Fine-Grained Inference Scaling of Structured Outputs for Function Calling
ToolPRM provides fine-grained intra-call process supervision via a new dataset and reward model, outperforming outcome and coarse-grained alternatives on function-calling benchmarks.
-
GeoLaux: A Benchmark for Evaluating MLLMs' Geometry Performance on Long-Step Problems Requiring Auxiliary Lines
GeoLaux is a new benchmark of 2186 long-step geometry problems requiring auxiliary lines, used to evaluate 23 MLLMs and reveal major drops in performance on complex tasks.
-
OSWorld-Human: Benchmarking the Efficiency of Computer-Use Agents
AI agents on OSWorld take 2.7-4.3 times more steps than human trajectories, with latency rising sharply due to repeated large model calls for planning and reflection.
-
R1-VL: Learning to Reason with Multimodal Large Language Models via Step-wise Group Relative Policy Optimization
R1-VL uses StepGRPO with rule-based StepRAR and StepRVR rewards to let MLLMs learn step-by-step reasoning beyond imitation of positive paths.
-
Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscation
Chain-of-thought monitoring detects reward hacking in frontier reasoning models, but strong optimization against the monitor produces obfuscated misbehavior that remains hard to detect.
-
Repair the Amplifier, Not the Symptom: Stable World-Model Correction for Agent Rollouts
WM-SAR identifies and repairs causal subgraphs that amplify errors in agent planning graphs, outperforming symptom-scanning LLM correctors under token constraints.
-
Path-level Hindsight Instructions for Semantic Exploration in Vision-Language Navigation
Phi-Nav generates path-level hindsight instructions from on-policy exploration trajectories to supply additional semantic supervision for vision-language navigation agents.
-
Benchmarking Large Language Models on Floating-Point Error Classification
Introduces InterFLOPBench benchmark and evaluates 14 LLMs on multi-label classification of six floating-point error categories in C code, with top models exceeding 0.88 overall F1 but lower scores on subtle errors like underflow.
-
Open Problems in Constitutional Preference Reconstruction
Empirical analysis across three datasets identifies three open problems in constitutional preference reconstruction and shows that principle refinement raises inter-executor agreement from 73% to 78%.
-
NormAct: A Benchmark for Hidden Social Norm Compliance in Embodied Planning
NormAct shows MLLMs reach explicit goals in 67.3% of cases but comply with hidden norms in only 26.4%, with NormPerceptor raising task success from 24.2% to 46.7%.
-
MER-R1: Multimodal Emotion Reasoning via Slow-Fast Thinking Synergy
MER-R1 uses dual-objective RL to optimize fast-thinking recall and slow-thinking precision separately in multimodal emotion recognition, with calibration to align them, yielding SOTA results on two benchmarks.
-
TOPS: First-Principles Visual Token Pruning via Constructing Token Optimal Preservation Sets for Efficient MLLM Inference
TOPS formulates visual token pruning as constructing Token Optimal Preservation Sets using three information-theoretic principles and demonstrates superior performance on MLLM benchmarks.
-
EHR-Complex: Benchmarking Medical Agents for Complex Clinical Reasoning
EHR-Complex is a new interactive benchmark on MIMIC-IV with 52K tasks averaging 31.93 SQL components, where top LLMs achieve 62.3% accuracy and exhibit SQL logic, medical-code, and semantic failures.
-
ThinkDeception: A Progressive Reinforcement Learning Framework for Interpretable Multimodal Deception Detection
ThinkDeception introduces MLLMs, a multimodal CoT dataset, and VAC-GRPO progressive RL to convert deception detection into interpretable reasoning and claims new SOTA accuracy plus rationale quality.