Beyond Inference-Only Deployment: Comparing Weight-Based Consolidation Against Cascading Compaction

• URL: http://arxiv.org/abs/2605.24657v1
• Date: Sat, 23 May 2026 16:57:25 GMT
• Title: Beyond Inference-Only Deployment: Comparing Weight-Based Consolidation Against Cascading Compaction
• Abstract summary: We evaluate a nightly consolidation of interaction knowledge into model weights via synthesis, synthesis, and Low-Rank Adaptation (LoRA) fine-tuning on a single consumer.<n>Across ten realistic software development conversations, cascading compaction retain 36.8 +/- 3.0% of knowledge.<n>LoRA consolidates interaction knowledge into model weights via synthesis, synthesis, and Low-Rank Adaptation (LoRA) fine-tuning on a single consumer.
• Score: 1.6214121783846343
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Major LLM platforms deploy models in an inference-only configuration: the model serves requests but never updates per-user weights. Users must repeatedly re-teach preferences, corrections, and project context, and context-based workarounds consume context-window space and degrade under cascading compaction. We evaluate an alternative: nightly consolidation of interaction knowledge into model weights via reflection, synthesis, and Low-Rank Adaptation (LoRA) fine-tuning on a single consumer GPU. Across ten realistic software development conversations (n = 10, 1,146 test questions across three memory types), three cycles of cascading compaction retain 36.8 +/- 3.0% of knowledge (between an 11.8% no-context floor and a 90.1% full-context ceiling), while consolidation retains 80.4 +/- 1.3% -- a 43.6 pp gain (paired t(9) = 14.8, p < 0.001) that more than doubles what compaction preserves, with the largest gains on procedural corrections (36.3% -> 74.6%) and episodic project facts (31.5% -> 78.2%). As a methodological aside, mean per-token validation cross-entropy is negatively correlated with LLM-judged accuracy (r = -0.51) while median per-token validation cross-entropy tracks accuracy almost exactly (r = +0.99): under evaluators that tolerate surface-form variation, the mean is misleading and a heavy-tail-robust statistic is the faithful signal. Persistent personalization requires moving beyond inference-only deployment toward architectures that consolidate knowledge into weights.

Related papers

• State of Thought Enables Endogenous Reasoning [48.902117018115256]
  Test-time reasoning has emerged as a major approach to improving the capabilities of Large Language Models (LLMs)<n>We propose State of Thought, a new reasoning paradigm that enables endogenous reasoning in LLMs.<n>We show that SoT consistently improves mean-baseline accuracy while reducing generated tokens by 62.6% and end-to-end latency by 44.6%.
  arXiv  Detail & Related papers  (2026-09-13T05:55:03Z)
• qZACH-ViT: Quantization-Aware Intrinsic Explanations with Recursive Attribution-Stabilized Optimization [0.0]
  We introduce qZACH-ViT, a quantization-aware extension of the zero-token-free ZACH-ViT backbone.<n>We also introduce Recursive Attribution-Stabilized Optimization (RASO), which norm-matches classification and attribution gradients.<n>We evaluate four controlled conditions on seven MedMNIST datasets using 50 training images per class and ten fixed seeds.
  arXiv  Detail & Related papers  (2026-07-16T19:51:48Z)
• Automated Proving of Shannon-Type Entropy Inequalities via Fine-Tuned Language Models and Guided Tree Search [50.16356451328644]
  Proving Shannon-type entropy inequalities is a fundamental task in information theory.<n>We investigate whether small-scale large language models, fine-tuned on atomic proof steps, can automate this process.<n>We show that GPT-5.5 solves 1.7% samples under zero-shot prompting while Psitip solves 33.3% samples.
  arXiv  Detail & Related papers  (2026-06-04T05:43:12Z)
• Models Can Model, But Can't Bind: Structured Grounding in Text-to-Optimization [54.749573452394664]
  We find that accuracy collapses as instance data grows, even when the formulation itself is simple.<n>We address this via a simple inference-time approach, BIND, which externalizes numeric data to structured files so the model binds data rather than from prompt prompt.<n>We validate our hypothesis by finetuning a model exclusively on binding and show that it outperforms end-to-end SFT and RL across three structurally distinct optimization categories.
  arXiv  Detail & Related papers  (2026-05-20T21:25:41Z)
• Amplifying, Not Learning: Fine-Tuned AI Text Detectors Amplify a Pretrained Direction [51.56484100374058]
  Text detectors amplify a pretrained typicality axis; they do not construct an AI-vs-human boundary.<n>On raw encoders before any task supervision, projecting onto centroid(AI)-centroid(HC3) achieves NYT-vs-HC3 AUROC 0.806/0.944/0.834 across three architectures.<n>On RoBERTa-base, raw projection exceeds fine-tuning; on RoBERTa-base, full fine-tuning reduces discrimination below raw on both fluent-formal populations tested.
  arXiv  Detail & Related papers  (2026-05-20T19:08:38Z)
• Metric Unreliability in Multimodal Machine Unlearning: A Systematic Analysis and Principled Unified Score [14.579552536669217]
  Unlearning in Vision-Language Models (VLMs) is required for compliance with the General Data Protection Regulation Code.<n>We present the first study of reliability in multimodal unlearning.<n>We introduce the Unified Quality Score (UQS), a metric with weights derived from each metric's Spearman correlation with the metric distance.
  arXiv  Detail & Related papers  (2026-05-04T04:13:00Z)
• Putting HUMANS first: Efficient LAM Evaluation with Human Preference Alignment [53.72927532626824]
  We show that subsets of just 50 examples (0.3% of data) can achieve over 0.93 Pearson correlation with full benchmark scores.<n>To better predict preferences, we trained regression models on these selected subsets, achieving 0.98 correlation.<n>This demonstrates that in regression modeling, well-curated subsets outpredict the full benchmark, showing quality over quantity.
  arXiv  Detail & Related papers  (2026-04-20T00:57:31Z)
• ACAR: Adaptive Complexity Routing for Multi-Model Ensembles with Auditable Decision Traces [3.151184728006369]
  We present ACAR, a measurement framework for studying multi-model orchestration under auditable conditions.<n>ACAR uses self-consistency variance (sigma) computed from N=3 probe samples to route tasks across single-model, two-model, and three-model execution modes.<n>We evaluate ACAR on 1,510 tasks spanning four benchmarks, producing more than 7,550 auditable runs.
  arXiv  Detail & Related papers  (2026-02-06T23:27:17Z)
• EdgeJury: Cross-Reviewed Small-Model Ensembles for Truthful Question Answering on Serverless Edge Inference [0.0]
  We present EdgeJury, a lightweight ensemble framework that improves truthfulness and robustness.<n>On TruthfulQA (MC1), EdgeJury achieves 76.2% accuracy.<n>On a 200-question adversarial EdgeCases set, EdgeJury yields +48.2% relative gains.
  arXiv  Detail & Related papers  (2025-12-29T14:48:40Z)
• Semantic-Constrained Federated Aggregation: Convergence Theory and Privacy-Utility Bounds for Knowledge-Enhanced Distributed Learning [0.0]
  We introduce Semantic-Constrained Federated Aggregation (SCFA), a theoretically-grounded framework incorporating domain knowledge constraints into distributed optimization.<n>We prove SCFA convergence rate O(1/sqrt(T) + rho) where rho represents constraint violation rate, establishing the first convergence theory for constraint-based federated learning.<n>We validate our framework on manufacturing predictive maintenance using Bosch production data with 1.18 million samples and 968 sensor features.
  arXiv  Detail & Related papers  (2025-12-12T04:29:29Z)
• Pearl: A Foundation Model for Placing Every Atom in the Right Location [52.35027831422145]
  We introduce Pearl, a foundation model for protein-ligand cofolding at scale.<n>Pearl establishes a new state-of-the-art performance in protein-ligand cofolding.<n>Pearl surpasses AlphaFold 3 and other open source baselines on the public Runs N' Poses and PoseBusters benchmarks.
  arXiv  Detail & Related papers  (2025-10-28T17:36:51Z)
• MixedNUTS: Training-Free Accuracy-Robustness Balance via Nonlinearly Mixed Classifiers [41.56951365163419]
  "MixedNUTS" is a training-free method where the output logits of a robust classifier are processed by nonlinear transformations with only three parameters. MixedNUTS then converts the transformed logits into probabilities and mixes them as the overall output. On CIFAR-10, CIFAR-100, and ImageNet datasets, experimental results with custom strong adaptive attacks demonstrate MixedNUTS's vastly improved accuracy and near-SOTA robustness.
  arXiv  Detail & Related papers  (2024-02-03T21:12:36Z)
• Patch-Level Contrasting without Patch Correspondence for Accurate and Dense Contrastive Representation Learning [79.43940012723539]
  ADCLR is a self-supervised learning framework for learning accurate and dense vision representation. Our approach achieves new state-of-the-art performance for contrastive methods.
  arXiv  Detail & Related papers  (2023-06-23T07:38:09Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.