Image de couverture de livre

Adaptive Task Balancing for Visual Instruction Tuning via Inter-Task Contribution and Intra-Task Difficulty

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Dai, Yanqi, Wang, Yong, You, Zebin, Jing, Dong, Chu, Xiangxiang, Lu, Zhiwu
Format:	Preprint
Publié:	2024
Sujets:	Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Harder Is Better: Boosting Mathematical Reasoning via Difficulty-Aware GRPO and Multi-Aspect Question Reformulation
par: Dai, Yanqi, et autres
Publié: (2026)

Bridging Writing Manner Gap in Visual Instruction Tuning by Creating LLM-aligned Instructions
par: Jing, Dong, et autres
Publié: (2025)

Investigating the Role of Instruction Variety and Task Difficulty in Robotic Manipulation Tasks
par: Parekh, Amit, et autres
Publié: (2024)

AnyTaskTune: Advanced Domain-Specific Solutions through Task-Fine-Tuning
par: Cui, Jiaxi, et autres
Publié: (2024)

Goldilocks RL: Tuning Task Difficulty to Escape Sparse Rewards for Reasoning
par: Mahrooghi, Ilia, et autres
Publié: (2026)

Where and What Matters: Sensitivity-Aware Task Vectors for Many-Shot Multimodal In-Context Learning
par: Ma, Ziyu, et autres
Publié: (2025)

TaskSense: Cognitive Chain Modeling and Difficulty Estimation for GUI Tasks
par: Yin, Yiwen, et autres
Publié: (2025)

Escaping Optimization Stagnation: Taking Steps Beyond Task Arithmetic via Difference Vectors
par: Wang, Jinping, et autres
Publié: (2025)

TaskEval: Assessing Difficulty of Code Generation Tasks for Large Language Models
par: Tambon, Florian, et autres
Publié: (2024)

LLaTiSA: Towards Difficulty-Stratified Time Series Reasoning from Visual Perception to Semantics
par: Ding, Yueyang, et autres
Publié: (2026)

Low-Resource Fine-Tuning for Multi-Task Structured Information Extraction with a Billion-Parameter Instruction-Tuned Model
par: Chih, Yu Cheng, et autres
Publié: (2025)

MMRole: A Comprehensive Framework for Developing and Evaluating Multimodal Role-Playing Agents
par: Dai, Yanqi, et autres
Publié: (2024)

Multi-Task Learning with Multi-Task Optimization
par: Bai, Lu, et autres
Publié: (2024)

HS-STaR: Hierarchical Sampling for Self-Taught Reasoners via Difficulty Estimation and Budget Reallocation
par: Xiong, Feng, et autres
Publié: (2025)

Generalization-Enhanced Code Vulnerability Detection via Multi-Task Instruction Fine-Tuning
par: Du, Xiaohu, et autres
Publié: (2024)

Dynamic Task Vector Grouping for Efficient Multi-Task Prompt Tuning
par: Zhang, Pieyi, et autres
Publié: (2025)

TCIA: A Task-Centric Instruction Augmentation Method for Instruction Finetuning
par: Ma, Simin, et autres
Publié: (2025)

Instruction Embedding: Latent Representations of Instructions Towards Task Identification
par: Li, Yiwei, et autres
Publié: (2024)

Parameter-Efficient Sparsity Crafting from Dense to Mixture-of-Experts for Instruction Tuning on General Tasks
par: Wu, Haoyuan, et autres
Publié: (2024)

Dual-Balancing for Multi-Task Learning
par: Lin, Baijiong, et autres
Publié: (2023)

Can Structured Templates Facilitate LLMs in Tackling Harder Tasks? : An Exploration of Scaling Laws by Difficulty
par: Yang, Zhichao, et autres
Publié: (2025)

Task-Agnostic Learning to Accomplish New Tasks
par: Zhang, Xianqi, et autres
Publié: (2022)

Task-Driven Exploration: Decoupling and Inter-Task Feedback for Joint Moment Retrieval and Highlight Detection
par: Yang, Jin, et autres
Publié: (2024)

TAGCOS: Task-agnostic Gradient Clustered Coreset Selection for Instruction Tuning Data
par: Zhang, Jipeng, et autres
Publié: (2024)

ESAinsTOD: A Unified End-to-End Schema-Aware Instruction-Tuning Framework for Task-Oriented Dialog Modeling
par: Teng, Dechuan, et autres
Publié: (2026)

CrossPT: Exploring Cross-Task Transferability through Multi-Task Prompt Tuning
par: Pouramini, Ahmad, et autres
Publié: (2025)

Reinforcing Real-world Service Agents: Balancing Utility and Cost in Task-oriented Dialogue
par: Gao, Ning, et autres
Publié: (2026)

Superpose Task-specific Features for Model Merging
par: Qiu, Haiquan, et autres
Publié: (2025)

Instruction Diversity Drives Generalization To Unseen Tasks
par: Zhang, Dylan, et autres
Publié: (2024)

Evaluating LLMs for Visualization Tasks
par: Khan, Saadiq Rauf, et autres
Publié: (2025)

Deep Reinforcement Learning with Task-Adaptive Retrieval via Hypernetwork
par: Jin, Yonggang, et autres
Publié: (2023)

Instruction-Tuned, but Not More Verifiable Instruction-Following: A Cross-Task Diagnosis for LoRA Adapters
par: Zou, Junyi
Publié: (2026)

Intent-conditioned and Non-toxic Counterspeech Generation using Multi-Task Instruction Tuning with RLAIF
par: Hengle, Amey, et autres
Publié: (2024)

Task-Specific Generative Dataset Distillation with Difficulty-Guided Sampling
par: Li, Mingzhuo, et autres
Publié: (2025)

Efficient Multi-Task Learning via Generalist Recommender
par: Wang, Luyang, et autres
Publié: (2025)

Decomposing the Basic Abilities of Large Language Models: Mitigating Cross-Task Interference in Multi-Task Instruct-Tuning
par: Wang, Bing, et autres
Publié: (2026)

Discovering Novel LLM Experts via Task-Capability Coevolution
par: Dai, Andrew, et autres
Publié: (2026)

Rubrics to Tokens: Bridging Response-level Rubrics and Token-level Rewards in Instruction Following Tasks
par: Xu, Tianze, et autres
Publié: (2026)

How Multimodal LLMs Solve Image Tasks: A Lens on Visual Grounding, Task Reasoning, and Answer Decoding
par: Yu, Zhuoran, et autres
Publié: (2025)

ReqBrain: Task-Specific Instruction Tuning of LLMs for AI-Assisted Requirements Generation
par: Habib, Mohammad Kasra, et autres
Publié: (2025)