LATEST FROM AAIG
News
Two papers have been accepted at NeurIPS 2026
Efficient Test-time Adaptation through Candidate Verification and Divergence Shifts Differentiable Bit-Widths: Co-optimizing Pruning and Quantization via SVD for Ultra-Efficient LLM Compression
A paper has been accepted at CoRL 2026
FlowVLA: Enabling World Dynamics Understanding through Pluggable 3D Flow Learning
A paper has been accepted at Applied Energy
One-Cycle Based Battery Remaining Useful Life Prediction by Continuous Wavelet Scale Compression with Deep Neural Networks
A paper has been accepted at BMVC 2026
Re-calibrated Contrastive Loss for Transformation-Aware Prompt Conditioning in Vision-Language Models
Three papers have been accepted at EMNLP 2026
Colla-Q: Toward Collaborative Experts in MoE Quantization via Minimax Precision Balancing Layer-wise Curriculum Learning for Efficient LLM Compression Train Overcomplete, Deploy Compact: Scaling Recovery Capacity for Structured LLM Pruning
SAIL paper accepted to Interspeech 2026
A SAIL paper was accepted to Interspeech 2026 as an oral presentation.
SAIL paper accepted to ICML 2026
SAIL announced the acceptance of a paper to ICML 2026.
SAIL paper accepted to Findings of ACL 2026
SAIL announced the acceptance of a paper to Findings of ACL 2026.
AAIG research output
Highlight Papers 2026
A quick view of regular conference papers currently listed in the AAIG archive.
View all publications ↗18papers
- EMNLP6
- NeurIPS3
- CoRL2
- ICML2
- ACL1
- BMVC1
- ECCV1
- ICDM1
- Interspeech1
ICRA 2019
SweepNet: Wide-baseline Omnidirectional Depth Estimation
ECCV 2018
DFT-based Transformation Invariant Pooling Layer for Visual Classification
NeurIPS Workshop 2026
SARAM: Sparse Adapter Allocation with Reuse and Merging for Lifelong Robot Learning
ICLR 2020
Generalized Convolutional Forest Networks for Domain Generalization and Visual Recognition
NeurIPS 2026
Differentiable Bit-Widths: Co-optimizing Pruning and Quantization via SVD for Ultra-Efficient LLM Compression
NeurIPS 2026
Efficient Test-time Adaptation through Candidate Verification and Divergence Shifts
CoRL 2026
FlowVLA: Enabling World Dynamics Understanding through Pluggable 3D Flow Learning
NeurIPS 2026
WaveGen: Truly End-to-End Waveform Generation via Internal Spectral Trajectory Learning
Research Areas
- MMAI Lab
- iKnow Lab
- LAMDA Lab
Vision & Perception
Reliable visual understanding, representation learning, and perception across real-world environments.
- MMAI Lab
- SAIL
- iKnow Lab
- LAMDA Lab
Multimodal Understanding
Methods that connect visual, audio, language, and structured information to reason across modalities.
- MMAI Lab
- HEI Lab
- iKnow Lab
- LAMDA Lab
Learning Systems & Foundation Models
Efficient, scalable, and generalizable learning systems for language, vision, and physical intelligence.
- SAIL
- iKnow Lab
Speech & Language AI
Speech synthesis, spoken-language modeling, translation, and knowledge-aware language intelligence.
- MMAI Lab
- SAIL
Generative AI
Generative methods for audio, video, visual content, and interactive multimodal media.
- MMAI Lab
- HEI Lab
Embodied Intelligence & Robot Learning
Vision-language-action learning and adaptable robot skills for interaction in the physical world.
- SAIL
- HEI Lab
- iKnow Lab
Human-Centered AI
Human-aware systems that collaborate, communicate, and adapt to people in practical settings.
- iKnow Lab
- LAMDA Lab
Knowledge, Recommendation & Personalization
Personalized and knowledge-centered intelligence for recommendation, retrieval, and assistance.
- MMAI Lab
- HEI Lab
- LAMDA Lab
Applied AI for Industry & Health
Robust AI methods for industrial inspection, biomedical data, and dependable real-world deployment.
- LAMDA Lab
- MMAI Lab
- iKnow Lab