5940 papers
This paper introduces Template-aware Dialogue Sentence Embedding (TaDSE), a novel self-supervised contrastive learning method that leverages easily obtainable token-level template information to generate high-quality sentence embeddings for task-oriented dialogues, achieving significant performance improvements over state-of-the-art methods on five benchmark datasets.
The paper introduces SciTune, a framework that aligns large language models with human-curated scientific multimodal instructions, resulting in a model (LLaMA-SciTune) that significantly outperforms state-of-the-art systems on scientific visual and language benchmarks, even surpassing human performance in certain categories.
The paper introduces MM-LIMA, a multimodal large language model that achieves superior performance compared to MiniGPT-4 by utilizing a trainable data selector to filter and fine-tune on a small, high-quality dataset of just 200 examples, demonstrating that less but higher-quality instruction data is more effective for alignment.
This paper proposes PredFT, a novel fMRI-to-text decoding framework that leverages predictive coding theory by integrating brain predictive representations from specific regions of interest into a main network, thereby outperforming existing models in reconstructing continuous language from brain signals.
This paper proposes ITEM, an iterative framework inspired by Schutz's philosophical system of relevances that enhances the retrieval, utility judgment, and answer generation stages of Retrieval-Augmented Generation (RAG) systems, demonstrating significant improvements across multiple datasets.
LaMI proposes a late multi-image fusion method that augments text-only Large Language Models with test-time visual signals generated via parallel sampling, significantly improving visual commonsense reasoning and NLP performance without the high costs of full multimodal training or the degradation of textual capabilities.
This paper introduces RiTeK, a comprehensive benchmark dataset designed to evaluate and improve Large Language Models' complex reasoning capabilities over medical Textual Knowledge Graphs by addressing current limitations in data scarcity, topological expressiveness, and retrieval performance.
This paper reveals that current LLM unlearning methods inadvertently create backdoor vulnerabilities by aligning forget-tokens with target representations, and proposes a model-agnostic Random Noise Augmentation (RNA) technique to mitigate this issue while preserving unlearning performance.
This paper introduces Ultra-Low-dimensional Prompt Tuning (ULPT), a parameter-efficient fine-tuning method that optimizes prompts in an ultra-low-dimensional space (e.g., 2D) using a frozen random up-projection matrix, achieving a 98% reduction in trainable parameters while outperforming existing methods across over 20 NLP tasks.
This paper introduces CounterBench, a new benchmark for evaluating large language models on formal rule-based counterfactual reasoning where they currently struggle, and proposes CoIn, an iterative reasoning paradigm that significantly improves their performance on these tasks.
This paper proposes AdaInit, a novel framework that leverages large language models with submartingale properties to iteratively synthesize adaptive initial parameters for Quantum Neural Networks, thereby effectively mitigating barren plateaus and ensuring higher gradient variance across various scales.
This paper introduces LIFT, a novel framework that enhances the long-context understanding of short-context large language models by dynamically fine-tuning model parameters to absorb long inputs via synthetic tasks, thereby enabling inference without the input context while avoiding quadratic complexity and maintaining low latency.
This paper addresses the scarcity of high-quality psycho-counseling data by proposing professional evaluation principles to create the 36k-pair PsychoCounsel-Preference dataset, which enables a fine-tuned Llama3-8B model to achieve an 87% win rate against GPT-4o in counseling tasks through reward modeling and preference learning.
The paper introduces OlymMATH, a rigorously curated Olympiad-level math benchmark featuring 350 manually sourced problems in dual English and Chinese versions that support both natural language and formal Lean 4 verification to address the saturation of existing evaluations and reveal persistent reasoning gaps in large language models.
This paper introduces LIFESTATE-BENCH, a novel benchmark utilizing narrative datasets like Hamlet to evaluate lifelong learning in large language models, revealing that while non-parametric methods outperform parametric ones in managing stateful interactions, all models still struggle with catastrophic forgetting over extended engagements.
This paper proposes Module Switching Defense (MSD), a novel post-training strategy that disrupts backdoor attacks by selectively fusing model modules via an evolutionary algorithm, demonstrating superior robustness and utility preservation compared to weight averaging, particularly in scenarios with limited models or collusive backdoors.
The paper introduces PODS, a method that accelerates reinforcement learning with verifiable rewards by decoupling rollout generation from policy updates and training only on a strategically down-sampled subset of rollouts, achieving the same performance as standard GRPO up to 1.7 times faster.
The paper proposes TokUR, a framework that estimates token-level uncertainty in Large Language Models using low-rank random weight perturbation to effectively assess and improve the reliability of mathematical reasoning responses.
This paper introduces DeceptionDecoded, a large-scale benchmark and intent-guided simulation framework designed to evaluate and improve vision-language models' ability to detect misleading creator intent in multimodal news, addressing their current reliance on superficial cues and enhancing their robustness in misinformation governance.
GoT-R1 is a novel framework that leverages reinforcement learning with a dual-stage multi-dimensional reward system to enhance the semantic-spatial reasoning capabilities of multimodal large language models, significantly improving their ability to generate images from complex prompts involving precise object relationships and attributes.