{"total":13,"items":[{"citing_arxiv_id":"2606.30382","ref_index":14,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"RQP: Resource-Oriented Quantiser Pruning for Neural Networks on FPGAs","primary_cat":"cs.AR","submitted_at":"2026-06-29T14:39:38+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"RQP reduces search cost up to 20.58x versus standard monotonic HGQ workflows on jet substructure classification while producing competitive Pareto frontiers for FPGA neural network accelerators.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.26538","ref_index":17,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"CascadeFormer: Depth-Tapered Transformers Motivated by Gradient Fan-in Asymmetry","primary_cat":"cs.LG","submitted_at":"2026-06-25T02:25:00+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"CascadeFormer tapers Transformer width with depth based on gradient fan-in asymmetry to match uniform baselines in perplexity while cutting latency 8.6%.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.22019","ref_index":22,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Channel Location Constrains the Auditability of Subliminal Learning","primary_cat":"cs.LG","submitted_at":"2026-06-20T12:48:31+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Auditability of subliminal learning is constrained by channel location, with initialization-dependent body channels allowing pre-training screens while vocabulary geometry and conditional body channels evade them.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.27479","ref_index":14,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Resource-Constrained Affect Modelling via Variance Regularisation Pruning","primary_cat":"cs.LG","submitted_at":"2026-05-26T12:05:13+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":3.0,"formal_verification":"none","one_line_summary":"Variance-Regularised Pruning maintains competitive CCC performance at 80% sparsity on the AGAIN dataset by incorporating cross-participant variance into the pruning process.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.18822","ref_index":2,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Hybrid-LoRA: Bridging Full Fine-Tuning and Low-Rank Adaptation for Post-Training","primary_cat":"cs.LG","submitted_at":"2026-05-12T15:11:44+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Hybrid-LoRA selectively full fine-tunes modules with high sensitivity to low-rank adaptation using a novel score and applies LoRA elsewhere, matching full fine-tuning at 10% budget and outperforming PEFT baselines by up to 5.65%.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.12207","ref_index":50,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Not How Many, But Which: Parameter Placement in Low-Rank Adaptation","primary_cat":"cs.LG","submitted_at":"2026-05-12T14:46:00+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Gradient-informed placement of LoRA parameters recovers full performance under GRPO while random placement does not, due to differences in gradient rank and stability across training regimes.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"SNIP: single-shot network pruning based on connection sensitivity. CoRR, abs/1810.02340, 2018. URLhttp://arxiv. org/abs/1810.02340. [49] Chunyuan Li, Heerad Farkhoor, Rosanne Liu, and Jason Yosinski. Measuring the intrinsic dimension of objective landscapes. In International Conference on Learning Representations, 2018. URLhttps://openreview.net/forum?id=ryup8-WCW. [50] Yang Li, Shaobo Han, and Shihao Ji. Vb-lora: Extreme parameter efficient fine-tuning with vector banks. In The 38th Conference on Neural Information Processing Systems (NeurIPS), 2024. [51] Ziniu Li, Tian Xu, Yushun Zhang, Zhihang Lin, Yang Yu, Ruoyu Sun, and Zhi-Quan Luo. Remax: a simple, effective, and efficient reinforcement learning method for aligning large"},{"citing_arxiv_id":"2605.09639","ref_index":11,"ref_count":2,"confidence":0.98,"is_internal_anchor":true,"paper_title":"XTinyU-Net: Training-Free U-Net Scaling via Initialization-Time Sensitivity","primary_cat":"eess.IV","submitted_at":"2026-05-10T16:34:39+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A Jacobian sensitivity curve computed at initialization identifies the narrowest U-Net configuration that avoids performance collapse, matching nnU-Net accuracy with 400-1600x fewer parameters on six medical datasets.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.08568","ref_index":34,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Different Prompts, Different Ranks: Prompt-aware Dynamic Rank Selection for SVD-based LLM Compression","primary_cat":"cs.LG","submitted_at":"2026-05-09T00:02:33+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"PARSE trains a prompt-aware linear router on dense-model outputs to select dynamic SVD ranks, improving accuracy up to 10% at 0.6 compression ratio on LLaMA-7B while delivering 2.5x prefill and 2.4x decode speedups.","context_count":1,"top_context_role":"method","top_context_polarity":"use_method","context_text":"combinations of calibration and evaluation datasets in Figure 2(a). Performance drops consistently when the evaluation dataset differs from the calibration set, with the largest gaps appearing when a specialized corpus such as MathQA or PTB is used for calibration, as reported in Figure 2(b). To diagnose the cause, we run the full SVD model without truncation and measure the rank-level connection sensitivity [34] of each rank component across inputs from different tasks (Figure 2(c)). The top-ranked components remain consistently important across datasets, while the importance of tail ranks is highly dataset-dependent. For WikiText-2, the critical components concentrate in the top ranks, whereas MathQA relies on a different and more dispersed subset; the asymmetry holds"},{"citing_arxiv_id":"2604.17396","ref_index":97,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Representation-Guided Parameter-Efficient LLM Unlearning","primary_cat":"cs.CL","submitted_at":"2026-04-19T11:59:58+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"REGLU guides LoRA-based unlearning via representation subspaces and orthogonal regularization to outperform prior methods on forget-retain trade-off in LLM benchmarks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.04988","ref_index":13,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Prune-Quantize-Distill: An Ordered Pipeline for Efficient Neural Network Compression","primary_cat":"cs.LG","submitted_at":"2026-04-05T06:13:47+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"The prune-quantize-distill ordering produces a better accuracy-size-latency frontier on CIFAR-10/100 than any single technique or other orderings, with INT8 QAT providing the main runtime gain.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2505.17469","ref_index":38,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"Efficient compression of neural networks and datasets","primary_cat":"cs.LG","submitted_at":"2025-05-23T04:50:33+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Refined probabilistic and smooth l0 pruning techniques approximate minimum description length for neural networks, achieving high compression with minimal accuracy loss and empirically verifying better sample efficiency and generalization on image and text tasks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2310.12508","ref_index":198,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation","primary_cat":"cs.LG","submitted_at":"2023-10-19T06:17:17+00:00","verdict":"CONDITIONAL","verdict_confidence":"MODERATE","novelty_score":6.0,"formal_verification":"none","one_line_summary":"SalUn uses gradient-based weight saliency to achieve effective machine unlearning of data, classes, or concepts in image classification and generation, narrowing the gap to exact retraining.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"1906.11626","ref_index":6,"ref_count":1,"confidence":0.98,"is_internal_anchor":true,"paper_title":"On improving deep learning generalization with adaptive sparse connectivity","primary_cat":"cs.NE","submitted_at":"2019-06-27T13:31:30+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Sparse MLPs trained via SET plus neuron pruning achieve competitive performance on 15 datasets while pruning ~50% of hidden neurons and keeping parameter count linear in neuron count.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null}],"limit":50,"offset":0}