MLLM framework detects hidden emotions with temporal context
The work builds a benchmark with expressed, internal emotion and concealment annotations to distinguish ordinary inconsistency from concealment using MLLM.
Stories per day · 30 days
2 RSS
Relatedaction-units1attribute-inference1emotion-detection1hidden-emotion1privacy1safe-llava1split-inference1temporal-context1
14 stories, newest first · one card per story
Stories tagged #multimodal-llm
Split MLLM features still leak age, gender, and race
With 20 labeled samples per class, Safe-LLaVA split features recover 75-85% of unprotected biometric leakage.
Rethinking Projector Training in Multimodal LLMs
PORTAL initializes multimodal projectors from summary statistics and matches caption-trained alignment without paired data.
Octrees as an Explicit 3D Language
OctLLM feeds 3D shapes to an LLM as sparse octree tokens via frozen-backbone trainable branches, cutting image-to-3D FID 17.4% versus ShapeLLM-Omni.
Reinforcing Multimodal Reasoning via Token-Level Perception-Grounded Advantage Estimation
TPAE uses token-level visual-dependency and entropy signals to improve multimodal RL reasoning over sequence-level rewards.
Imagine3D-LLM: Teaching MLLMs to Imagine 3D Scenes Before Answering
Imagine3D-LLM adds learnable summary tokens decoded into compact 3D Gaussian Splatting representations, improving multimodal LLM spatial reasoning across benchmarks.
Devils in Question Relay: Source-Conditioned Relay Steering to Mitigate Hallucinations in Audio-visual Large Language Models
SECRET, a training-free steering method, cuts source-confused hallucinations in audio-visual language models.
Dissecting Agentic Forensics: The Role of Triage, Prompting, and Evidence Arbitration in Open-World Fake Image Detection
Agentic fake-image forensics depends more on the MLLM judge's reasoning than on detector fusion.
Document Retrieval-Aware Chunking (D-RAC): Universal Retrieval-Aware Ingestion of Enterprise Documents via PDF Normalization and Multimodal Markdown Conversion
D-RAC converts enterprise documents into retrieval-ready Markdown chunks, cutting chunking tokens 95.7% versus agentic methods.
Not All Prompts Are Equal: Exploration-Guided Prompt Scaffolding for Multimodal Reinforcement Post-Training
Exploration-guided prompt scaffolding rewrites training prompts by Exploration Potential Score, boosting multimodal RL post-training accuracy up to 11.5%.
Anchoring Clinical Events in Time: UID-Preserving Multimodal Reconstruction and Source-Grounded Adjudication
UID-preserving multimodal framework plus GAVEL LLM judge improves clinical timeline reconstruction, boosting event recovery 43% over prior matching.
MindTopo: Can Foundation Models Reason in Topological Space?
MindTopo benchmark with 11,030 topological tasks shows 14 multimodal LLMs reason better than they plan and remain far below human performance.
ReactHuman: A Physics-Grounded Benchmark for Human-Like Reactive Decision-Making in Embodied Multimodal LLMs
ReactHuman benchmark tests whether multimodal LLMs react safely to sudden household hazards; seven evaluated models mishandle roughly one hazard in three.
VeriScene: Reconstructing Crime Scenes from Legal Evidence via World-Model Agent
Researchers present VeriScene, a world-model agent that reconstructs crime scenes from forensic photos and witness statements with traceable, physically plausible output.