{"arxiv_id":"2310.05193","incoming":{"papers":0,"sentences":0,"groups":[],"unlabelled_sentences":0},"outgoing":{"papers":20,"sentences":43,"groups":[{"kind":"method_or_tool","label":"Uses the method or tool of","papers":3,"sentences":3,"quotes":[{"arxiv_id":"1711.05101","title":"Decoupled Weight Decay Regularization","url":"/paper/1711.05101","section":"Experimental Settings","sentence":"When fine-tuning the language model in a uni-modal setting, we use BertAdam [ref] as the optimizer; We use SGD when training the optical flow in UCF101; For all other experiments, we employ AdamW [ref] .","intent_p":1.0,"contests_p":null},{"arxiv_id":"1810.04805","title":"BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding","url":"/paper/1810.04805","section":"Experimental Settings","sentence":"When fine-tuning the language model in a uni-modal setting, we use BertAdam [ref] as the optimizer; We use SGD when training the optical flow in UCF101; For all other experiments, we employ AdamW [ref] .","intent_p":1.0,"contests_p":null},{"arxiv_id":"2104.01778","title":"AST: Audio Spectrogram Transformer","url":"/paper/2104.01778","section":"Experimental Settings","sentence":"Our audio preprocessing is consistent with that of [ref] or [ref] ; (3) For text, we use the corresponding tokenizer for different language models.","intent_p":0.97,"contests_p":null}]},{"kind":"data_used","label":"Uses data or values from","papers":1,"sentences":1,"quotes":[{"arxiv_id":"1212.0402","title":"UCF101: A Dataset of 101 Human Actions Classes From Videos in The Wild","url":"/paper/1212.0402","section":"Datasets","sentence":"The UCF101 dataset, presented by [ref] , is designed for action recognition, featuring 101 distinct action categories.","intent_p":0.84,"contests_p":null}]},{"kind":"comparison","label":"Compares with","papers":2,"sentences":2,"quotes":[{"arxiv_id":"1909.02950","title":"Supervised Multimodal Bitransformers for Classifying Images and Text","url":"/paper/1909.02950","section":"Main Results of MMLoRA","sentence":"Here, we compare MMLoRA with MBT [ref] , MMBT [ref] , and PMF [ref] .","intent_p":1.0,"contests_p":null},{"arxiv_id":"2305.01233","title":"On Uni-Modal Feature Learning in Supervised Multi-Modal Learning","url":"/paper/2305.01233","section":"Main Results of MMLoRA","sentence":"The methods compared include balancing training by adding loss (G-Blending [ref] , PMR [ref] , UMT [ref] ), adjusting the training progress of different modalities by modifying gradients (OGM-GE [ref] ), novel fusion framework (MBT [ref] ),or directly averaging uni-modal predictions (UME [ref] ).","intent_p":0.98,"contests_p":null}]},{"kind":"supports_this_paper","label":"Cites as support","papers":1,"sentences":2,"quotes":[{"arxiv_id":"2305.01233","title":"On Uni-Modal Feature Learning in Supervised Multi-Modal Learning","url":"/paper/2305.01233","section":"Multi-Modal Low-Rank Adaptation learning (MMLoRA)","sentence":"As can be seen from Section 3.2 of the paper [ref] , in multi-modal datasets, different modalities need to adapt to each other to produce better results.","intent_p":0.97,"contests_p":null}]},{"kind":"background","label":"Cites as background","papers":9,"sentences":10,"quotes":[{"arxiv_id":"2305.14314","title":"QLoRA: Efficient Finetuning of Quantized LLMs","url":"/paper/2305.14314","section":"Related Work","sentence":"Parameter-Efficient Fine-Tuning is a methodology designed to fine-tune only a subset of parameters in large language models, enhancing their adaptability to downstream tasks [ref] .","intent_p":0.99,"contests_p":null},{"arxiv_id":"2106.09685","title":"LoRA: Low-Rank Adaptation of Large Language Models","url":"/paper/2106.09685","section":"Related Work","sentence":"Parameter-Efficient Fine-Tuning is a methodology designed to fine-tune only a subset of parameters in large language models, enhancing their adaptability to downstream tasks [ref] .","intent_p":0.99,"contests_p":null},{"arxiv_id":"2110.07602","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","url":"/paper/2110.07602","section":"Related Work","sentence":"Parameter-Efficient Fine-Tuning is a methodology designed to fine-tune only a subset of parameters in large language models, enhancing their adaptability to downstream tasks [ref] .","intent_p":0.99,"contests_p":null},{"arxiv_id":"2303.10512","title":"AdaLoRA: Adaptive Budget Allocation for Parameter-Efficient Fine-Tuning","url":"/paper/2303.10512","section":"Related Work","sentence":"Parameter-Efficient Fine-Tuning is a methodology designed to fine-tune only a subset of parameters in large language models, enhancing their adaptability to downstream tasks [ref] .","intent_p":0.99,"contests_p":null},{"arxiv_id":"2304.08485","title":"Visual Instruction Tuning","url":"/paper/2304.08485","section":"Introduction","sentence":"By connecting pre-trained language and vision models, MLLMs equip language models with the ability to `see' [ref] .","intent_p":0.81,"contests_p":null}]}],"unlabelled_sentences":25},"claims":[],"thresholds":{"intent_p":0.8,"supports_p":0.95,"contests_p":0.7},"method":"Citing sentences come from each paper's LaTeX source; every \\cite marks the sentence and the works it cites. Pith's classifier typed each sentence. A label is shown at intent probability 0.8 or more (0.95 for support), a contest at 0.7 or more. Measured accuracy against two blind model families is at https://pith.science/accuracy. The quoted sentences are the citing authors' own.","held_up":null,"_links":{"html":"https://pith.science/paper/2310.05193"}}