cs.LGNov 1, 2026

Same Function, Different Gradients: Evaluating Gradient Reliability under Reparameterization

Authors: Bum Jun Kim

Organizations: The University of Tokyo

Abstract

Gradients are no longer used only to train models: Gradient magnitudes are now measurement instruments, used to rank attention heads for pruning, to decide per-example clipping in private training, to allocate adapter rank budgets, and to debug models. Across these workflows, larger gradient magnitudes are taken to mean greater importance. Even for a fixed input--output function, these gradients are computed in one of many possible parameter coordinate systems. For example, attention scores depend on the query and key projections only through the paired product WQWK⊤W_QW_K^\top. Scaling a head's query projection by cc and the corresponding key projection by 1/c1/c therefore leaves every output logit unchanged. This change in parameter coordinates rescales the corresponding query and key gradient norms by 1/c1/c and cc, respectively. To test whether gradient-based decisions remain unchanged under these coordinate changes, we introduce Gradient Reliability under Reparameterization (GRR). GRR rewrites a trained model in new parameter coordinates that compute exactly the same function, verifies that predictions are unchanged, and then tests whether the decisions derived from the gradients are unchanged too. Our audits span GLUE classifiers, a ViT on CIFAR-10, GPT-2-medium and GPT-2-large, a 1.1B-parameter Llama-style model on WikiText-2, and six trained LoRA adapters. Empirically, predictions are preserved throughout, but both the gradients and the gradient-based decisions change. Across 66 GLUE attention audits, the median top-3 overlap of raw head-gradient rankings is only 1/31/3, and rankings and pruning masks change even under the mildest rescalings we test. In the largest of 14 global-pruning audits, the shifted ranking selects head masks that differ in four of 22 pruned heads and reach 50.92%50.92\% and 94.84%94.84\% validation accuracy on one ELECTRA SST-2 model.

Figures & tables

Appendix figures & tables18 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

Jun 3, 2026

GRAIL: Gradient-Reweighted Advantages for Reinforcement Learning with Verifiable Rewards

Reinforcement learning with verifiable rewards (e.g. GRPO) is now a common way to improve mathematical reasoning in Large Language Models (LLMs). However, current methods usually broadcast one sequence-level advantage to all tokens, or use costly process reward models (PRMs) for step-level supervision. Uniform advantage distribution assumes that all tokens contribute equally to the final reward. This dilutes the gradient signal, since flawed reasoning steps and filler words are updated as strongly as valid logical inferences. To address this, we introduce Gradient-Reweighted Advantage (GRAIL), an intrinsic token-wise advantage reweighting method. GRAIL uses gradient-activation saliency to place more weight on tokens that are more locally sensitive to the final answer. Evaluations across five models from the Qwen3, R1-distilled and OctoThinker families show that GRAIL consistently outperforms GRPO. GRAIL achieved an average improvement of 3.60% in accuracy and 3.05% in Pass@3, demonstrating that fine-grained reasoning alignment can be achieved without process-level supervision.
Oct 14, 2026

ReGRA: Residual-Gradient Adaptation for Complementary Low-Rank Fine-Tuning

Low-rank adaptation (LoRA) enables parameter-efficient fine-tuning by learning low-rank weight updates. Increasing rank expands adaptation capacity, but does not guarantee that all learned directions improve downstream performance. In a diagnostic experiment with rank-16 LoRA, we show that retaining only the top singular components achieves the best performance, whereas adding further components degrades it. Motivated by this, we propose Residual-Gradient Adaptation (ReGRA), a two-stage method for making more effective use of a fixed rank budget. After the first stage, ReGRA estimates task gradients on a calibration set and removes their projections onto the two-sided subspace induced by the first-stage update. It then aggregates the residual-gradient directions to initialize a second-stage adapter without changing the first-stage update. An orthogonality regularizer encourages the two updates to capture complementary directions. Both updates can be merged into the pretrained weights, incurring no additional inference overhead. Across diverse benchmark sets, ReGRA achieves stronger performance than alternatives with the same total rank and even matches LoRA adapters with twice the rank. Ablation studies and further analysis support the contribution of its core design choices. Code is available at https://github.com/Fanyao-Li/ReGRA.
Jun 14, 2026

Retrievable Gradients: Continual Post-Training Without Cumulative Weight Drift

Continual post-training enables models to absorb emerging knowledge after deployment, but repeatedly updating shared parameters can accumulate weight drift, potentially causing catastrophic forgetting and degrading general capabilities. Retrieval-augmented generation avoids such parameter drift, yet often lacks the depth of parametric knowledge integration. In this paper, we propose ReGrad (Retrievable Gradients), a new paradigm that treats gradients as retrievable units of knowledge. ReGrad pre-computes document-specific gradients offline, stores them in an indexed Gradient Bank, and retrieves only query-relevant gradients at inference time for temporary weight adaptation. However, raw language-modeling gradients are optimized for token-level document reconstruction rather than for query-driven knowledge use. We therefore introduce a bi-level meta-learning objective that reshapes document-derived gradients into generalizable adaptation signals for downstream tasks. Experiments across general and domain-specific settings show that ReGrad outperforms CPT and RAG baselines, enabling scalable and reversible parametric knowledge injection without accumulating weight drift.