Posters

Researchers unveil text-level graph injection attacks, revealing a new vulnerability in GNNs and highlighting the importance of text interpretability in attack success.

Robots using LLMs for task planning often make unsafe or wrong decisions due to LLM hallucination and ambiguity in instructions. This paper introduces ‘introspective planning,’ a novel method that us…

Spectral Attention boosts long-range dependency capture in time series forecasting, achieving state-of-the-art results across various models and datasets.

Strategic players can manipulate reward signals, but simple threshold policies still achieve a surprisingly good approximation to the optimal prophet value, even in this more realistic setting.

IntraMix: Boost GNN accuracy by cleverly generating high-quality labels and enriching node neighborhoods using intra-class Mixup.

This paper introduces a novel framework for creating interventionally consistent surrogate models for complex simulations, addressing computational limitations and ensuring accurate policy evaluation.

This study presents CADIM, an adaptive algorithm using interventions to learn true causal relationships from mixtures of DAGs, achieving near-optimal intervention sizes and providing quantifiable opti…

Researchers uniquely estimate probabilities in Causal Bayesian Networks using simple independence assumptions, enabling analysis from observational data and simplifying counterfactual probability calc…

Researchers model a manifold of customized diffusion models as a subspace of weights, enabling controllable creation of new models via sampling, editing, and inversion from a single image.

Researchers developed methods to measure and interpret the divergence between learned feedback patterns (LFPs) in LLMs and human preferences, helping minimize discrepancies between LLM behavior and tr…

SpLiCE unlocks CLIP’s potential by transforming its dense, opaque representations into sparse, human-interpretable concept embeddings.

CLIP’s zero-shot image classification decisions are made interpretable using a novel mutual-knowledge approach based on textual concepts, demonstrating effective and human-friendly analysis across div…

Interpretable Mesomorphic Neural Networks (IMNs) achieve accuracy comparable to black-box models while offering free-lunch explainability for tabular data through instance-specific linear models gener…

Interpretable lightweight transformers are built by unrolling graph smoothness priors, achieving high performance with significantly fewer parameters than conventional transformers.

ProtoViT: a novel interpretable image classification method using Vision Transformers and adaptive prototypes, achieving higher accuracy and providing clear explanations.

M-GAM: Interpretable additive models handling missing data with superior accuracy & sparsity!

CMR: A novel Concept-Based Memory Reasoner delivers human-understandable, verifiable AI task predictions by using a neural selection mechanism over a set of human-understandable logic rules, achievin…

Successive Concept Bottleneck Agents (SCoBots) improve reinforcement learning by integrating interpretable layers, enabling concept-level inspection and human-in-the-loop revisions to fix misalignment…

Problem (FAIR) framework and FORM algorithm achieve flexible multi-stakeholder fairness in online recommendation systems, balancing platform revenue with user and item fairness.

InternLM-XComposer2-4KHD pioneers high-resolution image understanding in LVLMs, scaling processing from 336 pixels to 4K HD and beyond, achieving state-of-the-art results on multiple benchmarks.