Posters

New computational method infers complex brain learning rules from experimental data, revealing active forgetting in reward learning.

Mobility-LLM leverages LLMs to analyze human mobility data from check-in sequences, significantly outperforming existing models in location prediction, user identification, and time prediction tasks.

Mobile-Agent-v2 uses a three-agent collaborative framework (planning, decision, reflection) to improve mobile device operation accuracy by over 30%, overcoming the limitations of single-agent architec…

MO-DDN: A new benchmark and coarse-to-fine exploration agent boosts embodied AI’s ability to handle multi-object, preference-based task planning.

MMSite: a novel multi-modal framework accurately identifies protein active sites using protein sequences and textual descriptions, achieving state-of-the-art performance.

AVMoE: a novel parameter-efficient transfer learning approach for audio-visual learning, dynamically allocates expert models (unimodal and cross-modal adapters) based on task demands, achieving superi…

Mixture of Tokens (MoT) achieves 3x faster LLM training than dense Transformers and matches state-of-the-art MoE performance via continuous token mixing.

BinaryMoS: a novel token-adaptive binarization method that boosts LLM accuracy and efficiency by dynamically merging multiple scaling experts for each token.

Hydra: a novel cryo-EM reconstruction method resolves both conformational and compositional heterogeneity ab initio, enabling the analysis of complex, unpurified samples with state-of-the-art accuracy…

Link-MoE boosts link prediction accuracy by strategically selecting the best model for each node pair, surpassing single-model approaches.

MoICE, a novel plug-in, significantly enhances LLMs’ long context awareness by dynamically routing attention using multiple RoPE angles, achieving superior performance with high inference efficiency.

Non-linear Residual Gates (NoRGa) boosts prompt-based continual learning by theoretically framing prefix tuning as adding new experts to a pre-trained Mixture-of-Experts model, achieving state-of-the-…

MoD, a novel Mixture of Demonstrations framework, enhances in-context learning by partitioning demonstration pools and employing expert-wise training, achieving state-of-the-art performance.

Boosting Robust Few-Shot Learning with Adversarial Meta-Tuning!

MixEval revolutionizes LLM benchmarking by blending real-world user queries with existing datasets, creating a cost-effective, unbiased, and dynamic evaluation method.

A new formula unifies lazy and active neural network training regimes, revealing a mixed regime that combines their strengths for faster convergence and low-rank bias.

DPR, a novel bias mitigation method, robustly improves model performance by leveraging disagreement probability without needing bias labels, achieving state-of-the-art results.

ADVPO, a novel method, tackles reward overoptimization in RLHF via a lightweight uncertainty quantification approach, resulting in enhanced LLM performance and alignment with human values.

New metric, λ-discrepancy, precisely detects & mitigates partial observability in sequential decision processes, significantly boosting reinforcement learning agent performance.

Concentric Causal Attention (CCA) significantly reduces object hallucination in LVLMs by cleverly reorganizing visual tokens to mitigate the impact of long-term decay in Rotary Position Encoding.