Poster sessions

Poster Session I (Morning session)

1 ImageNet-RIB Benchmark: Large Pre-Training Datasets Don't Guarantee Robustness after Fine-Tuning

2 Best Unpacking DPO and PPO: Disentangling Practices for Learning from Preference Feedback

3 Ensembling Finetuned Language Models for Text Classification

4 One Initialization to Rule them All: Fine-tuning via Explained Variance Adaptation

5 CRAFT Your Dataset: Task-Specific Synthetic Dataset Generation Through Corpus Retrieval and Augmentation

6 Understanding Visual Concepts Across Models

7 A Layer Selection Approach to Test Time Adaptation

8 ActNAS : Generating Efficient YOLO Models using Activation NAS

9 On Efficient Distillation from LLMs to SLMs

10 Towards Exploring Continual Fine-Tuning for Enhancing Language Ability in Large Language Model

11 Fine tuning language models to align fidelity and efficiency of generative retrieval in multi-turn dialogues

12 PAL: Pluralistic Alignment Framework for Learning from Heterogeneous Preferences

13 A Tensor-based Convolutional Neural Network for Small Dataset Classification

14 Generalizing Alignment Paradigm of Text-to-Image Generation with Preferences through $f$-divergence Minimization

15 Parasite Networks: Transfer Learning in Resource-Constrained Domains

16 Flat-LoRA: Low-Rank Adaption over a Flat Loss Landscape

17 Estimating Effects of Tokens in Preference Learning

18 Efficient Fine-Tuning of CNN-based Foundation Models for Segmentation in 3D Medical Images

19 FourierKAN outperforms MLP on Text Classification Head Fine-tuning

20 E-Tamba: Efficient Transformer-Mamba Layer Transplantation

21 Inducing Semi-Structured Sparsity by Masking for Efficient Model Inference in Convolutional Networks

22 Token Pruning using a Lightweight Background Aware Vision Transformer

23 Improving LLM Generation with Inverse and Forward Alignment: Reward Modeling, Prompting, Fine-Tuning, and Inference-Time Optimization

24 Inconsistencies In Consistency Models: Better ODE Solving Does Not Imply Better Samples

25 Faster, More Efficient RLHF through Off-Policy Asynchronous Learning

26 Comparing Bad Apples to Good Oranges: Aligning Large Language Models via Joint Preference Optimization

27 Online Fine-Tuning with Uncertainty Quantification for Offline Pre-Trained Agents

28 Addax: Resource-Efficient Fine-Tuning of Language Models with a Combination of Forward-Backward and Forward-Only Passes

29 Learning the Regularization Strength for Deep Fine-Tuning via a Data-Emphasized Variational Objective

30 Towards Long-Context Time Series Foundation Models With A Handful Of Additional Parameters

31 What Causes a Disparate Impact in a Quantized Model?

32 Investigating the Role of Fine-Tuning in Addressing the Gap Between Synthetic and Real Data in Generative Foundation Models

33 Model Soup for Better RLHF: Weight Space Averaging to Improve Alignment in LLMs

34 Balancing Cost and Effectiveness of Synthetic Data Generation Strategies for LLMs

35 Flexora: Flexible Low-Rank Adaptation for Large Language Models

36 REACT: Residual-Adaptive Contextual Tuning for Fast Model Adaptation in Cybersecurity

37 Adapting Language Models via Token Translation

38 Efficient Fine-Tuning of Behavior Cloned Policies with Reinforcement Learning from Limited Demonstrations

39 Towards Natural Machine Unlearning

40 Self-Stitching: Widely Applicable and Efficient Transfer Learning Using Stitching Layer

41 MPLoRA: Orthogonal Multi-Path Low-Rank Adaptation for Parameter Efficient Fine-Tuning

42 An empirical study of CLIP fine-tuning with similarity clusters

43 XoRA: Expander Adapted LoRA Finetuning

44 Early Exiting in Deep Neural Networks via Dirichlet-based Uncertainty Quantification

Poster Session II (Afternoon session)

1 Noise Stability Optimization for Finding Flat Minima: A Hessian-based Regularization Approach2 HyperDPO: Hypernetwork-based Multi-Objective Fine-Tuning Framework3 Analysing Softmax Entropy Minimization for Adaptating Multitask Models at Test-time4 Fine-tuning Vision Classifiers On A Budget5 Variational Low-Rank Adaptation Using IVON6 Mastering Task Arithmetic: $\tau$Jp as a Key Indicator for Weight Disentanglement7 DistRL: An Asynchronous Distributed Reinforcement Learning Framework for On-Device Control Agent8 TOU: Truncated-factorized reduction for an efficient-parameter model fine-tuning9 Discrepancy-Guided Parameter Suppression for Robust Fine-tuning10 Scalability of memorization-based machine unlearning11 Memory retaining finetuning via distillation12 Improving Fine-Tuning with Latent Cluster Correction13 TreeTop: Topology-Aware Fine-Tuning for LLM Conversation Tree Understanding14 Uncertainty-Penalized Direct Preference Optimization15 FRACTAL: Fine-Grained Scoring from Aggregate Text Labels16 Variational Best-of-N Alignment17 Semi-Supervised Fine-Tuning of Vision Foundation Models with Content-Style Decomposition18 Sharp Analysis for KL-Regularized Contextual Bandits and RLHF19 Fitness Aware Human Motion Generation with Fine-Tuning20 Effective Text-to-Image Alignment with Quality Aware Pair Ranking21 GaLore-mini: Low Rank Gradient Learning with Fewer Learning Rates22 On the Transferability of Parameter-Efficient Continual Learning for Vision Transformers23 Hierarchical Unlearning Framework for Multi-Class Classification24 Accelerating Direct Preference Optimization with Prefix Sharing25 Enhancing Cross-Language Code Translation via Task-Specific Embedding Alignment in Retrieval-Augmented Generation26 FedEx-LoRA: Exact Aggregation for Federated Parameter-Efficient Fine-Tuning of Foundation Models27 Instant Transformer Adaption via HyperLoRA28 Learning Robust Representations for Transfer in Reinforcement Learning29 PLMFit: Benchmarking Transfer Learning with Protein Language Models for Protein Engineering30 Simultaneous Weight and Architecture Optimization for Neural Networks31 Optimizing Small Language Models for In-Vehicle Function-Calling32 Skip Transformers: Efficient Inference through Skip-Routing33 SVFT: Parameter-Efficient Fine-Tuning with Singular Vectors34 Evaluating Fine-Tuning Efficiency of Human-Inspired Learning Strategies in Medical Question Answering35 Unintentional Unalignment: Likelihood Displacement in Direct Preference Optimization36 LLM Alignment Through Successive Policy Re-weighting (SPR)37 Teaching LLMs How To Learn with Contextual Fine-Tuning38 Deep Reinforcement Learning Without Experience Replay, Target Networks, or Batch Updates39 Instruct-SkillMix: A Powerful Pipeline for LLM Instruction Tuning40 Learning Reward and Policy Jointly from Demonstration and Preference Improves Alignment41 Navigating Parameter Space with Geodesic Interpolation: A New Approach to Efficient Fine-Tuning42 Characterizing the Training Dynamics of Private Fine-tuning with Langevin diffusion43 UnoLoRA: Single Low-Rank Adaptation for Efficient Multitask Fine-tuning

Page updated

Google Sites

Report abuse