EHRNote-ChatQA is the first benchmark for evidence-grounded multi-turn clinical QA over longitudinal discharge summaries, containing 16,072 medical-expert-verified pairs across eight categories and revealing LLM weaknesses in evidence grounding and multi-turn consistency.
super hub Mixed citations
Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
Mixed citation behavior. Most common role is background (55%).
961
Pith papers citing it
Background
55% of classified citations
abstract
In this report, we introduce the Gemini 2.X model family: Gemini 2.5 Pro and Gemini 2.5 Flash, as well as our earlier Gemini 2.0 Flash and Flash-Lite models. Gemini 2.5 Pro is our most capable model yet, achieving SoTA performance on frontier coding and reasoning benchmarks. In addition to its incredible coding and reasoning skills, Gemini 2.5 Pro is a thinking model that excels at multimodal understanding and it is now able to process up to 3 hours of video content. Its unique combination of long context, multimodal and reasoning capabilities can be combined to unlock new agentic workflows. Gemini 2.5 Flash provides excellent reasoning abilities at a fraction of the compute and latency requirements and Gemini 2.0 Flash and Flash-Lite provide high performance at low latency and cost. Taken together, the Gemini 2.X model generation spans the full Pareto frontier of model capability vs cost, allowing users to explore the boundaries of what is possible with complex agentic problem solving.
hub tools
citation-role summary
background 122
baseline 46
method 28
other 8
dataset 3
citation-polarity summary
claims ledger
- abstract In this report, we introduce the Gemini 2.X model family: Gemini 2.5 Pro and Gemini 2.5 Flash, as well as our earlier Gemini 2.0 Flash and Flash-Lite models. Gemini 2.5 Pro is our most capable model yet, achieving SoTA performance on frontier coding and reasoning benchmarks. In addition to its incredible coding and reasoning skills, Gemini 2.5 Pro is a thinking model that excels at multimodal understanding and it is now able to process up to 3 hours of video content. Its unique combination of long context, multimodal and reasoning capabilities can be combined to unlock new agentic workflows. G
authors
co-cited works
representative citing papers
HKJudge is a new ~290k-sentence expert-annotated corpus of Hong Kong criminal judgments with 26 rhetorical roles and 3 sentencing elements, plus benchmarks on classification and extraction tasks.
Introduces the first longitudinal voice dataset for RRP with benchmarks across handcrafted features, deep networks, self-supervised models, and audio LLMs under patient-level validation.
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.
EgoIntrospect provides the first egocentric dataset with self-annotations for internal state tasks and shows multimodal LLMs struggle to infer subjective states from combined signals.
Persona vectors form within the first 0.22% of LLM pretraining and remain effective for steering post-trained models, with continued refinement and transfer to other models.
Sieve dynamically schedules MoE experts across GPU and PIM hardware to handle bimodal token distributions, achieving 1.3x to 1.6x gains in throughput and interactivity over static prior PIM systems on three large models.
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.
Omni-DeepSearch is a 640-sample benchmark for audio-driven omni-modal search where the best model reaches only 43.44% accuracy, exposing bottlenecks in audio inference, tool use, and cross-modal 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.
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.
MM-JudgeBench shows substantial cross-lingual performance variance in 22 LVLM judges, with model size and architecture as poor predictors of multilingual robustness.
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.
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.
Large language models display the identifiable victim effect at roughly twice the human baseline, strongly amplified by instruction tuning and chain-of-thought prompting but inverted by reasoning-specialized models.
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.
HM-Bench is the first benchmark for MLLMs on hyperspectral images, showing models struggle with complex spatial-spectral reasoning and perform better with visual PCA images than textual reports.
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.
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.
ScreenParse dataset and ScreenVLM model deliver dense screen parsing that outperforms larger VLMs on PageIoU and transfers to better UI grounding.
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.
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.
ConceptPose delivers state-of-the-art zero-shot relative pose estimation by matching open-vocabulary 3D concept vectors derived from VLM saliency maps, beating the strongest baseline by 62% in ADD(-S) without training.
citing papers explorer
-
AutoPyVerifier: Learning Compact Executable Verifiers for Large Language Model Outputs
AutoPyVerifier learns compact sets of executable Python verifiers from labeled LLM outputs via LLM synthesis and DAG search, improving objective prediction by up to 55 F1 points and downstream LLM accuracy by up to 17 points.
-
TTS-PRISM: A Perceptual Reasoning and Interpretable Speech Model for Fine-Grained Diagnosis
TTS-PRISM defines a 12-dimensional perceptual schema, builds a targeted diagnostic dataset via adversarial synthesis and expert labels, and tunes an end-to-end model that outperforms generalist LLMs in human alignment on a 1,600-sample Mandarin test set while profiling six TTS paradigms.
-
Seeing Fast and Slow: Learning the Flow of Time in Videos
Self-supervised models learn to perceive and manipulate the flow of time in videos, supporting speed detection, large-scale slow-motion data curation, and temporally controllable video synthesis.
-
Long-Horizon Manipulation via Trace-Conditioned VLA Planning
LoHo-Manip enables robust long-horizon robot manipulation by using a receding-horizon VLM manager to output progress-aware subtask sequences and 2D visual traces that condition a VLA executor for automatic replanning.
-
Adaptive Test-Time Compute Allocation with Evolving In-Context Demonstrations
An adaptive test-time framework uses a warm-up phase on the test set to build evolving in-context examples, then concentrates compute on unresolved queries to outperform static baselines on math, coding, and reasoning tasks with lower total inference cost.
-
OMIBench: Benchmarking Olympiad-Level Multi-Image Reasoning in Large Vision-Language Model
OMIBench benchmark reveals that current LVLMs achieve at most 50% on Olympiad problems requiring reasoning across multiple images.
-
WebGen-R1: Incentivizing Large Language Models to Generate Functional and Aesthetic Websites with Reinforcement Learning
WebGen-R1 uses end-to-end RL with scaffold-driven generation and cascaded rewards for structure, function, and aesthetics to transform a 7B model into a generator of deployable multi-page websites that rivals much larger models.
-
Building a Precise Video Language with Human-AI Oversight
CHAI framework pairs AI pre-captions with expert human critiques to produce precise video descriptions, enabling open models to outperform closed ones like Gemini-3.1-Pro and improve fine-grained control in video generation models.
-
Temporally Extended Mixture-of-Experts Models
Temporally extended MoE layers using the option-critic framework with deliberation costs cut switching rates below 5% while retaining most capability on MATH, MMLU, and MMMLU.
-
An Agentic Approach to Metadata Reasoning
Metadata Reasoner uses agentic LLM reasoning on metadata to select sufficient and minimal data sources, achieving 83.16% F1 on KramaBench and 85.5% F1 on noisy synthetic benchmarks while avoiding low-quality tables 99% of the time.
-
Infection-Reasoner: A Compact Vision-Language Model for Wound Infection Classification with Evidence-Grounded Clinical Reasoning
Infection-Reasoner, a 4B VLM, reaches 86.8% accuracy on wound infection classification while producing rationales rated mostly correct by experts, via GPT-5.1 distillation followed by reinforcement learning.
-
SafetyALFRED: Evaluating Safety-Conscious Planning of Multimodal Large Language Models
SafetyALFRED shows multimodal LLMs recognize kitchen hazards accurately in QA tests but achieve low success rates when required to mitigate those hazards through embodied planning.
-
Thinking Before Matching: A Reinforcement Reasoning Paradigm Towards General Person Re-Identification
ReID-R achieves competitive person re-identification performance using chain-of-thought reasoning and reinforcement learning with only 14.3K non-trivial samples, about 20.9% of typical data scales, while providing interpretations.
-
SAW-INT4: System-Aware 4-Bit KV-Cache Quantization for Real-World LLM Serving
Token-wise INT4 KV-cache quantization plus block-diagonal Hadamard rotation recovers nearly all accuracy lost by naive INT4 while adding zero end-to-end overhead under paged serving constraints.
-
Detoxification for LLM: From Dataset Itself
HSPD detoxifies pretraining corpora via hierarchical semantic-preserving rewriting with Soft Contrastive Decoding, cutting toxicity probability from 0.42 to 0.18 and expected maximum toxicity from 0.43 to 0.20 on GPT2-XL with consistent gains on other models.
-
KD-Judge: A Knowledge-Driven Automated Judge Framework for Functional Fitness Movements on Edge Devices
KD-Judge structures fitness rules via LLM retrieval and chain-of-thought, then uses pose-guided kinematics for rule-based rep validation with caching for efficient edge deployment, achieving RTF < 1 and speedups up to 15.91x on Jetson.
-
OpenGame: Open Agentic Coding for Games
OpenGame is the first open-source agentic framework for end-to-end web game creation, using Game Skills and GameCoder-27B to achieve state-of-the-art results on 150 prompts via a new benchmark measuring build health, visual usability, and intent alignment.
-
MFMDQwen: Multilingual Financial Misinformation Detection Based on Large Language Model
MFMDQwen is the first open-source LLM for multilingual financial misinformation detection, backed by a new instruction dataset and benchmark on which it outperforms other open-source models.
-
Model in Distress: Sentiment Analysis on French Synthetic Social Media
A backtranslation-based synthetic data pipeline produces 1.7 million French tweets to train reasoners that reach 77-79% accuracy on human-annotated distress detection, matching or beating proprietary LLMs.
-
Training LLM Agents for Spontaneous, Reward-Free Self-Evolution via World Knowledge Exploration
LLM agents trained with a task-success reward on self-generated knowledge can spontaneously explore and adapt to new environments without any rewards or instructions at inference, yielding 20% gains on web tasks and allowing a 14B model to beat Gemini-2.5-Flash.
-
Unmasking the Illusion of Embodied Reasoning in Vision-Language-Action Models
State-of-the-art vision-language-action models catastrophically fail dynamic embodied reasoning due to lexical-kinematic shortcuts, behavioral inertia, and semantic feature collapse caused by architectural bottlenecks, as shown by the new BeTTER benchmark with real-world validation.
-
LoReC: Rethinking Large Language Models for Graph Data Analysis
LoReC enhances LLMs for graph tasks via attention redistribution, graph re-injection into FFN, and logit rectification, yielding improvements over GraphLLM and GNN baselines on diverse datasets.
-
VIBE: Voice-Induced open-ended Bias Evaluation for Large Audio-Language Models via Real-World Speech
VIBE evaluates generative biases in large audio-language models with real-world speech and open-ended tasks, showing that gender cues produce larger distributional shifts than accent cues across 11 tested models.
-
Beyond Text-Dominance: Understanding Modality Preference of Omni-modal Large Language Models
Omni-modal LLMs exhibit visual preference that emerges in mid-to-late layers, enabling hallucination detection without task-specific training.
-
Expressing Social Emotions: Misalignment Between LLMs and Human Cultural Emotion Norms
Frontier LLMs over-express engaging emotions relative to disengaging ones and generate deterministic responses that fail to match the cultural and individual diversity observed in human social emotion expression.
-
Scalable and Adaptive Parallel Training of Graph Transformer on Large Graphs
A new distributed framework for graph transformer training auto-selects parallel strategies and optimizes sparse operations to deliver up to 6x speedup on 8 GPUs and 78% memory reduction.
-
AVRT: Audio-Visual Reasoning Transfer through Single-Modality Teachers
AVRT transfers reasoning to audio-visual models by distilling traces from single-modality teachers via LLM merger followed by SFT cold-start and RL, achieving SOTA on OmniBench, DailyOmni, and MMAR with 3B/7B models.
-
Repurposing 3D Generative Model for Autoregressive Layout Generation
LaviGen turns 3D generative models into an autoregressive layout generator that models geometric and physical constraints, delivering 19% higher physical plausibility and 65% faster inference on the LayoutVLM benchmark.
-
LLMs Corrupt Your Documents When You Delegate
LLMs corrupt an average of 25% of document content during long delegated editing workflows across 52 domains, even frontier models, and agentic tools do not mitigate the issue.
-
MM-WebAgent: A Hierarchical Multimodal Web Agent for Webpage Generation
MM-WebAgent is a hierarchical multimodal agent that coordinates AIGC tools through planning and iterative self-reflection to generate coherent, visually consistent webpages and outperforms baselines on a new benchmark.
-
ControlFoley: Unified and Controllable Video-to-Audio Generation with Cross-Modal Conflict Handling
ControlFoley introduces a unified framework for controllable video-to-audio generation using joint visual encoding, temporal-timbre decoupling, and robust multimodal training to handle cross-modal conflicts.
-
Listen, Pause, and Reason: Toward Perception-Grounded Hybrid Reasoning for Audio Understanding
HyPeR is a hybrid perception-reasoning framework that uses a new hierarchical PAQA dataset and PAUSE tokens to improve large audio language models' handling of multi-speaker and ambiguous audio.
-
Development of an LLM-Based System for Automatic Code Generation from HEP Publications
A two-stage LLM system extracts structured analysis selections from HEP papers and references then generates and validates executable code, achieving partial event-level matches on an ATLAS Higgs-to-four-leptons benchmark but limited by hallucination and stochasticity.
-
MARCA: A Checklist-Based Benchmark for Multilingual Web Search
MARCA is a bilingual benchmark using 52 questions and validated checklists to evaluate LLM web-search completeness and correctness in English and Portuguese.
-
Decoding the Delta: Unifying Remote Sensing Change Detection and Understanding with Multimodal Large Language Models
Delta-LLaVA adds Change-Enhanced Attention, Change-SEG with prior embeddings, and Local Causal Attention to MLLMs to overcome temporal blindness, outperforming general models on a new unified benchmark for bi- and tri-temporal remote sensing tasks.
-
Evaluation of Agents under Simulated AI Marketplace Dynamics
Marketplace Evaluation uses repeated-interaction simulations to assess information access systems with marketplace-level metrics such as retention and market share that complement traditional accuracy measures.
-
UHR-BAT: Budget-Aware Token Compression Vision-Language model for Ultra-High-Resolution Remote Sensing
UHR-BAT is a budget-aware framework that uses text-guided multi-scale importance estimation plus region-wise preserve and merge strategies to compress visual tokens in ultra-high-resolution remote sensing vision-language models.
-
TOPCELL: Topology Optimization of Standard Cell via LLMs
TOPCELL reformulates standard cell topology optimization as an LLM generative task with GRPO fine-tuning, outperforming base models and matching exhaustive solvers with 85.91x speedup in 2nm/7nm industrial flows.
-
SceneCritic: A Symbolic Evaluator for 3D Indoor Scene Synthesis
SceneCritic is a symbolic, ontology-grounded evaluator for floor-plan layouts that identifies specific semantic, orientation, and geometric violations and aligns better with human judgments than VLM-based scorers.
-
SpotSound: Enhancing Large Audio-Language Models with Fine-Grained Temporal Grounding
SpotSound adds a hallucination-suppressing objective and a needle-in-haystack benchmark to audio-language models, reaching state-of-the-art temporal grounding while keeping general task performance.
-
Grasp in Gaussians: Fast Monocular Reconstruction of Dynamic Hand-Object Interactions
GraG reconstructs dynamic 3D hand-object interactions from monocular video 6.4x faster than prior work by using compact Sum-of-Gaussians tracking initialized from large models and refined with 2D losses.
-
MetFuse: Figurative Fusion between Metonymy and Metaphor
MetFuse provides the first dataset of 1,000 meaning-aligned quadruplets fusing literal, metonymic, metaphoric, and hybrid sentences, which augments training to boost metonymy and metaphor classification performance on benchmarks.
-
Don't Show Pixels, Show Cues: Unlocking Visual Tool Reasoning in Language Models via Perception Programs
Perception Programs rewrite dense visual tool outputs into language-native summaries, boosting MLLM accuracy by 15-45% absolute on BLINK perception tasks and setting new state-of-the-art results.
-
MISID: A Multimodal Multi-turn Dataset for Complex Intent Recognition in Strategic Deception Games
MISID is a multimodal multi-turn dataset for intent recognition in strategic deception games, paired with the FRACTAM framework that improves MLLM performance on hidden intent detection via decouple-anchor-reason steps.
-
TEMPLATEFUZZ: Fine-Grained Chat Template Fuzzing for Jailbreaking and Red Teaming LLMs
TEMPLATEFUZZ mutates chat templates with element-level rules and heuristic search to reach 98.2% average jailbreak success rate on twelve open-source LLMs while degrading accuracy by only 1.1%.
-
MLLM-as-a-Judge Exhibits Model Preference Bias
MLLMs show self-preference bias and family-level mutual bias when judging captions; Philautia-Eval quantifies it and Pomms ensemble reduces it.
-
Seeing Through Touch: Tactile-Driven Visual Localization of Material Regions
The model uses dense visuo-tactile feature interactions and material-diversity pairing on expanded datasets to generate tactile saliency maps for material segmentation, outperforming prior global-alignment methods.
-
Bridging What the Model Thinks and How It Speaks: Self-Aware Speech Language Models for Expressive Speech Generation
SA-SLM uses variational information bottleneck for intent-aware bridging and self-criticism for realization-aware alignment to close the semantic-acoustic gap, outperforming open-source models and nearing GPT-4o-Audio expressiveness on EchoMind after training on 800 hours of data.
-
The Salami Slicing Threat: Exploiting Cumulative Risks in LLM Systems
Salami Attack chains low-risk inputs to cumulatively trigger high-risk LLM behaviors, achieving over 90% success on GPT-4o and Gemini while resisting some defenses.
-
Erasing Thousands of Concepts: Towards Scalable and Practical Concept Erasure for Text-to-Image Diffusion Models
ETC scales concept erasure to thousands of concepts in T2I diffusion models via tMM modeling, affine optimal transport, and a robust MoEraser module.