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.
hub Mixed citations
Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs
Mixed citation behavior. Most common role is background (56%).
67
Pith papers citing it
Background
56% of classified citations
abstract
We introduce Phi-4-Mini and Phi-4-Multimodal, compact yet highly capable language and multimodal models. Phi-4-Mini is a 3.8-billion-parameter language model trained on high-quality web and synthetic data, significantly outperforming recent open-source models of similar size and matching the performance of models twice its size on math and coding tasks requiring complex reasoning. This achievement is driven by a carefully curated synthetic data recipe emphasizing high-quality math and coding datasets. Compared to its predecessor, Phi-3.5-Mini, Phi-4-Mini features an expanded vocabulary size of 200K tokens to better support multilingual applications, as well as group query attention for more efficient long-sequence generation. Phi-4-Multimodal is a multimodal model that integrates text, vision, and speech/audio input modalities into a single model. Its novel modality extension approach leverages LoRA adapters and modality-specific routers to allow multiple inference modes combining various modalities without interference. For example, it now ranks first in the OpenASR leaderboard to date, although the LoRA component of the speech/audio modality has just 460 million parameters. Phi-4-Multimodal supports scenarios involving (vision + language), (vision + speech), and (speech/audio) inputs, outperforming larger vision-language and speech-language models on a wide range of tasks. Additionally, we experiment to further train Phi-4-Mini to enhance its reasoning capabilities. Despite its compact 3.8-billion-parameter size, this experimental version achieves reasoning performance on par with or surpassing significantly larger models, including DeepSeek-R1-Distill-Qwen-7B and DeepSeek-R1-Distill-Llama-8B.
hub tools
citation-role summary
background 10
baseline 4
dataset 1
other 1
citation-polarity summary
claims ledger
- abstract We introduce Phi-4-Mini and Phi-4-Multimodal, compact yet highly capable language and multimodal models. Phi-4-Mini is a 3.8-billion-parameter language model trained on high-quality web and synthetic data, significantly outperforming recent open-source models of similar size and matching the performance of models twice its size on math and coding tasks requiring complex reasoning. This achievement is driven by a carefully curated synthetic data recipe emphasizing high-quality math and coding datasets. Compared to its predecessor, Phi-3.5-Mini, Phi-4-Mini features an expanded vocabulary size of
co-cited works
representative citing papers
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.
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.
VideoOdyssey is a new benchmark featuring ultra-long videos (avg. 109 min) across 11 domains with multi-level continuous certificates (avg. 16 min for visual, 12.8 min for audio-visual) to diagnose MLLM limitations in continuous reasoning and omni-modal perception.
X-Token proposes projection-guided P-KL and H-KL losses to fix uncommon-token suppression and over-conservative matching in logit-based cross-tokenizer distillation, yielding gains over GOLD on Llama-3.2-1B.
CopT reverses CoT by eliciting a draft answer first then using continuous-embedding contrastive verification and on-policy thinking to reflect and correct, yielding up to 23% higher accuracy and 57% fewer tokens without training.
OmniPro is the first benchmark jointly evaluating omni-modal perception, proactive responding, and diverse streaming video understanding tasks using a dual-mode protocol on 2700 samples.
PAREDA is a new multi-accent speech dataset of spontaneous NLP paper discussions that shows state-of-the-art ASR models struggle in zero-shot settings but improve after fine-tuning on it.
CBT-Audio dataset shows that adding audio input improves distress intensity estimation over transcripts alone for 8 of 10 audio language models, with clearest gains when verbal content and vocal delivery diverge.
MetaBackdoor shows that LLMs can be backdoored using positional triggers like sequence length, enabling stealthy activation on clean inputs to leak system prompts or trigger malicious behavior.
Omni-Persona benchmark with 18 tasks shows open-source models have audio-visual grounding gaps, RLVR narrows them but leads to conservative outputs, and scale or recall alone fail as diagnostics.
Malicious Skills induce coding agents to hallucinate and import attacker-controlled packages at high rates while evading detection.
DAPRO provides the first dynamic, theoretically guaranteed way to allocate interaction budgets across test cases for bounding time-to-event in multi-turn LLM evaluations, achieving tighter coverage than static conformal survival methods.
RobotEQ is the first benchmark for active intelligence in embodied AI, demonstrating that current models underperform on social norm adherence and spatial grounding tasks.
Symmetric Nucleus Subsampling and Expert Embedding Engine reduce modality gaps in multimodal embeddings by over 90% and outperform baselines in data curation for downstream models.
SENECA uses a novel self-consistent missing mass calculation to improve discrete entropy estimates in small-sample regimes and outperforms alternatives in numerical tests.
A new multi-accent long-form call-center dialogue dataset for English ASR evaluation shows substantial performance variation across accents and segmentation methods.
Semantic-level and verification-based uncertainty methods outperform token-level baselines for audio reasoning in ALLMs, but their relative performance on hallucination and unanswerable-question benchmarks is model- and task-dependent.
Introduces MUSCAT benchmark dataset of bilingual scientific discussions to evaluate multilingual ASR performance on code-switching and mixed inputs beyond standard WER.
AudioHijack generates imperceptible adversarial audio via gradient estimation, attention supervision, and reverberation blending to hijack 13 LALMs with 79-96% success on unseen contexts and real commercial agents.
Agreeableness in AI personas reliably predicts sycophantic behavior in 9 of 13 tested language models.
GeoMMBench reveals deficiencies in current multimodal LLMs for geoscience tasks while GeoMMAgent demonstrates that tool-integrated agents achieve significantly higher performance.
OmniTrace converts token-level signals into span-level cross-modal attributions for open-ended generation in omni-modal LLMs via generation-time tracing.
AQUA-Bench evaluates audio QA models on three unanswerability scenarios: missing correct answers, mismatched choice sets, and questions irrelevant to the audio.
citing papers explorer
-
SenseBench: A Benchmark for Remote Sensing Low-Level Visual Perception and Description in Large Vision-Language Models
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.
-
HalluAudio: A Comprehensive Benchmark for Hallucination Detection in Large Audio-Language Models
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.
-
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.
-
VideoOdyssey: A Benchmark for Ultra-Long-Context and Omni-Modal Video Understanding
VideoOdyssey is a new benchmark featuring ultra-long videos (avg. 109 min) across 11 domains with multi-level continuous certificates (avg. 16 min for visual, 12.8 min for audio-visual) to diagnose MLLM limitations in continuous reasoning and omni-modal perception.
-
X-Token: Projection-Guided Cross-Tokenizer Knowledge Distillation
X-Token proposes projection-guided P-KL and H-KL losses to fix uncommon-token suppression and over-conservative matching in logit-based cross-tokenizer distillation, yielding gains over GOLD on Llama-3.2-1B.
-
CopT: Contrastive On-Policy Thinking with Continuous Spaces for General and Agentic Reasoning
CopT reverses CoT by eliciting a draft answer first then using continuous-embedding contrastive verification and on-policy thinking to reflect and correct, yielding up to 23% higher accuracy and 57% fewer tokens without training.
-
OmniPro: A Comprehensive Benchmark for Omni-Proactive Streaming Video Understanding
OmniPro is the first benchmark jointly evaluating omni-modal perception, proactive responding, and diverse streaming video understanding tasks using a dual-mode protocol on 2700 samples.
-
PAREDA: A Multi-Accent Speech Dataset of Natural Language Processing Research Discussions
PAREDA is a new multi-accent speech dataset of spontaneous NLP paper discussions that shows state-of-the-art ASR models struggle in zero-shot settings but improve after fine-tuning on it.
-
CBT-Audio: Evaluating Audio Language Models for Patient-Side Distress Intensity Estimation in CBT Session Recordings
CBT-Audio dataset shows that adding audio input improves distress intensity estimation over transcripts alone for 8 of 10 audio language models, with clearest gains when verbal content and vocal delivery diverge.
-
MetaBackdoor: Exploiting Positional Encoding as a Backdoor Attack Surface in LLMs
MetaBackdoor shows that LLMs can be backdoored using positional triggers like sequence length, enabling stealthy activation on clean inputs to leak system prompts or trigger malicious behavior.
-
Omni-Persona: Systematic Benchmarking and Improving Omnimodal Personalization
Omni-Persona benchmark with 18 tasks shows open-source models have audio-visual grounding gaps, RLVR narrows them but leads to conservative outputs, and scale or recall alone fail as diagnostics.
-
Trust Me, Import This: Dependency Steering Attacks via Malicious Agent Skills
Malicious Skills induce coding agents to hallucinate and import attacker-controlled packages at high rates while evading detection.
-
How Many Iterations to Jailbreak? Dynamic Budget Allocation for Multi-Turn LLM Evaluation
DAPRO provides the first dynamic, theoretically guaranteed way to allocate interaction budgets across test cases for bounding time-to-event in multi-turn LLM evaluations, achieving tighter coverage than static conformal survival methods.
-
RobotEQ: Transitioning from Passive Intelligence to Active Intelligence in Embodied AI
RobotEQ is the first benchmark for active intelligence in embodied AI, demonstrating that current models underperform on social norm adherence and spatial grounding tasks.
-
Multimodal Data Curation Through Ranked Retrieval
Symmetric Nucleus Subsampling and Expert Embedding Engine reduce modality gaps in multimodal embeddings by over 90% and outperform baselines in data curation for downstream models.
-
SENECA: Small-Sample Discrete Entropy Estimation via Self-Consistent Missing Mass
SENECA uses a novel self-consistent missing mass calculation to improve discrete entropy estimates in small-sample regimes and outperforms alternatives in numerical tests.
-
AppTek Call-Center Dialogues: A Multi-Accent Long-Form Benchmark for English ASR
A new multi-accent long-form call-center dialogue dataset for English ASR evaluation shows substantial performance variation across accents and segmentation methods.
-
Walking Through Uncertainty: An Empirical Study of Uncertainty Estimation for Audio-Aware Large Language Models
Semantic-level and verification-based uncertainty methods outperform token-level baselines for audio reasoning in ALLMs, but their relative performance on hallucination and unanswerable-question benchmarks is model- and task-dependent.
-
MUSCAT: MUltilingual, SCientific ConversATion Benchmark
Introduces MUSCAT benchmark dataset of bilingual scientific discussions to evaluate multilingual ASR performance on code-switching and mixed inputs beyond standard WER.
-
Hijacking Large Audio-Language Models via Context-Agnostic and Imperceptible Auditory Prompt Injection
AudioHijack generates imperceptible adversarial audio via gradient estimation, attention supervision, and reverberation blending to hijack 13 LALMs with 79-96% success on unseen contexts and real commercial agents.
-
Too Nice to Tell the Truth: Quantifying Agreeableness-Driven Sycophancy in Role-Playing Language Models
Agreeableness in AI personas reliably predicts sycophantic behavior in 9 of 13 tested language models.
-
GeoMMBench and GeoMMAgent: Toward Expert-Level Multimodal Intelligence in Geoscience and Remote Sensing
GeoMMBench reveals deficiencies in current multimodal LLMs for geoscience tasks while GeoMMAgent demonstrates that tool-integrated agents achieve significantly higher performance.
-
OmniTrace: A Unified Framework for Generation-Time Attribution in Omni-Modal LLMs
OmniTrace converts token-level signals into span-level cross-modal attributions for open-ended generation in omni-modal LLMs via generation-time tracing.
-
AQUA-Bench: Beyond Finding Answers to Knowing When There Are None in Audio Question Answering
AQUA-Bench evaluates audio QA models on three unanswerability scenarios: missing correct answers, mismatched choice sets, and questions irrelevant to the audio.
-
MCGA: A Multi-task Classical Chinese Literary Genre Audio Corpus
MCGA is a new 119-hour multi-task audio corpus for classical Chinese literary genres that shows current MLLMs face substantial challenges on its test set.
-
Hearing to Translate: The Effectiveness of Speech Modality Integration into LLMs
Cascaded systems remain the most reliable for speech translation overall, but recent SpeechLLMs match or outperform them in many conditions while standalone speech models lag.
-
TRANSPORTER: Transferring Visual Semantics from VLM Manifolds
TRANSPORTER generates videos from VLM logits using optimal transport to interpret model predictions on object attributes, actions, and scenes.
-
Audio Flamingo 3: Advancing Audio Intelligence with Fully Open Large Audio Language Models
Audio Flamingo 3 introduces an open large audio-language model achieving new state-of-the-art results on over 20 audio understanding and reasoning benchmarks using a unified encoder and curriculum training on open data.
-
OCRBench v2: An Improved Benchmark for Evaluating Large Multimodal Models on Visual Text Localization and Reasoning
OCRBench v2 is a new benchmark with four times more tasks than prior versions that reveals most large multimodal models score below 50 out of 100 on visual text tasks and share five specific weaknesses.
-
Artificial Intolerance: Stigmatizing Language in Clinical Documentation Skews Large Language Model Decision-Making
Frontier LLMs exhibit bias from stigmatizing language in clinical vignettes across four conditions, skewing decisions toward less aggressive management, with limited mitigation from Chain-of-Thought or self-debiasing prompts.
-
From Text to Voice: A Reproducible and Verifiable Framework for Evaluating Tool Calling LLM Agents
A dataset-agnostic framework converts text tool-calling benchmarks to paired audio evaluations via TTS, speaker variation and noise, then evaluates seven omni-modal models showing model- and task-dependent performance with small text-to-voice gaps.
-
Verification Mirage: Mapping the Reliability Boundary of Self-Verification in Medical VQA
Self-verification in medical VQA creates a verification mirage where verifiers exhibit high error and agreement bias on wrong answers, with reliability strongly conditioned on task type.
-
SimCT: Recovering Lost Supervision for Cross-Tokenizer On-Policy Distillation
SimCT enlarges the supervision space in cross-tokenizer on-policy distillation using short jointly tokenizable multi-token continuations, producing consistent gains over shared-token baselines on math and code benchmarks.
-
Experience Sharing in Mutual Reinforcement Learning for Heterogeneous Language Models
Mutual Reinforcement Learning allows heterogeneous LLMs to exchange experience through mechanisms like Peer Rollout Pooling, Cross-Policy GRPO Advantage Sharing, and Success-Gated Transfer, with outcome-level sharing identified as favorable on the stability-support trade-off.
-
HumanNet: Scaling Human-centric Video Learning to One Million Hours
HumanNet is a 1M-hour human-centric video dataset with interaction annotations that enables better vision-language-action model performance than equivalent robot data in a controlled test.
-
All That Glitters Is Not Audio: Rethinking Text Priors and Audio Reliance in Audio-Language Evaluation
Audio-language models retain 60-72% of benchmark scores without audio, and most audio-dependent items can be solved from short fragments rather than full clips.
-
HeadRouter: Dynamic Head-Weight Routing for Task-Adaptive Audio Token Pruning in Large Audio Language Models
HeadRouter prunes audio tokens more effectively by dynamically routing based on per-head importance for semantic versus acoustic tasks, exceeding baseline performance at 70% token retention on Qwen2.5-Omni models.
-
COMPASS: COntinual Multilingual PEFT with Adaptive Semantic Sampling
COMPASS uses semantic clustering on multilingual embeddings to select auxiliary data for PEFT adapters, outperforming linguistic-similarity baselines on multilingual benchmarks while supporting continual adaptation.
-
In-Situ Behavioral Evaluation for LLM Fairness, Not Standardized-Test Scores
Standardized-test benchmarks for LLM fairness are unreliable because prompt wording alone drives most score variance and ranking changes, while a multi-agent conversational framework reveals consistent model-specific fairness behaviors across millions of dialogues.
-
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.
-
GroupDPO: Memory efficient Group-wise Direct Preference Optimization
GroupDPO decouples group-wise preference optimization during backpropagation to cut peak memory while keeping the same gradients, allowing larger groups and consistent gains over single-pair DPO plus an NLL term on positives.
-
Contextual Biasing for ASR in Speech LLM with Common Word Cues and Bias Word Position Prediction
Common-word acoustic cues and bias-word position prediction in speech LLMs cut rare-word transcription errors by 16.3% versus baselines, including out-of-domain cases.
-
CheeseBench: Evaluating Large Language Models on Rodent Behavioral Neuroscience Paradigms
CheeseBench is a benchmark where LLMs act as zero-shot agents in text-rendered versions of classical rodent experiments, with the best model reaching 52.6% success compared to 32.1% random and 78.9% approximate rodent baselines.
-
Differences in Text Generated by Diffusion and Autoregressive Language Models
DLMs exhibit lower n-gram entropy, higher semantic coherence, and higher semantic diversity than ARMs, primarily due to bidirectional context and remasking decoding strategies.
-
Beyond Content Safety: Real-Time Monitoring for Reasoning Vulnerabilities in Large Language Models
An external zero-shot monitor detects nine unsafe reasoning behaviors in LLMs at 87% step-level accuracy with low false positives and low latency.
-
GroupGPT: A Token-efficient and Privacy-preserving Agentic Framework for Multi-User Chat Assistant
GroupGPT decouples intervention timing from response generation via edge-cloud collaboration for multi-user chats, scoring 4.72/5 on the new MUIR benchmark of 2500 segments while cutting token use by up to 3x and adding privacy sanitization.
-
PAL*M: Property Attestation for Large Generative Models
PAL*M is a property attestation framework for large generative models that combines confidential virtual machines, security-aware GPUs, and incremental multiset hashing to achieve low-overhead integrity tracking with formal security guarantees.
-
Genie Envisioner: A Unified World Foundation Platform for Robotic Manipulation
Genie Envisioner unifies robotic policy learning, simulation, and evaluation inside one instruction-conditioned video diffusion framework using GE-Base, GE-Act, and GE-Sim.
-
The Ratchet Effect in Silico through Interaction-Driven Cumulative Intelligence in Large Language Models
Populations of 1-4B parameter LLMs using peer verification and shared cultural memory achieve 8.8-18.9 point gains on mathematical reasoning tasks and close much of the gap to 70B+ single models.
-
LamPO: A Lambda Style Policy Optimization for Reasoning Language Models
LamPO introduces a pairwise decomposed advantage with confidence-aware weighting to replace scalar group advantages in group-relative policy optimization for reasoning models.