A Classifier That Teaches Itself: Self-Improving, Frozen-gate Training (SIFT) for Dynamic Document Classification

• URL: http://arxiv.org/abs/2607.18358v1
• Date: Mon, 20 Jul 2026 12:38:50 GMT
• Title: A Classifier That Teaches Itself: Self-Improving, Frozen-gate Training (SIFT) for Dynamic Document Classification
• Abstract summary: We present SIFT (Self-Improving, Frozen-gate Training), a dynamic classifier service.<n>We describe the architecture, the self-feeding corpus loop, the frozen-gate promotion mechanism, and an illustrative multi-domain deployment.
• Score: 39.17006317116773
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Document classification is a solved problem in the laboratory and an unsolved one in the enterprise. The blocker is rarely model architecture; it is the labeling project that must precede a model and the institutional fear of letting a model retrain itself once one exists. We present SIFT (Self-Improving, Frozen-gate Training), a dynamic classifier service, which attacks both. SIFT serves classification from a deliberately cheap, CPU-bound pipeline, a SPLADE sparse encoder feeding a LightGBM head, and escalates only the low-confidence minority of pages to an LLM judge. The judge's verdicts are written back into a labeled corpus, so the expensive model continuously teaches the cheap one: the escalation rate falls, the corpus grows from production traffic rather than from an up-front annotation effort, and accuracy compounds with use. Onboarding a new document family requires only a declarative bundle, label space, anchor phrases, and a judge glossary, not a labeling project. The harder problem is safety: an autonomously retraining classifier can silently regress. SIFT resolves this with a two-part promote gate, a critical-label F1 regression check plus a frozen golden regression set the model is never trained on, either of which vetoes promotion. This turns "retrain monthly without a human" from reckless into routine. We describe the architecture, the self-feeding corpus loop, the frozen-gate promotion mechanism, and an illustrative multi-domain deployment, and we discuss the economics of a classifier whose marginal labeling cost trends toward zero.

Related papers

• Fast Unlearning at Scale via Margin Self-Correction [52.46927918952516]
  Language-model unlearning updates a trained model to behave as if it had not seen selected training examples.<n>MASC achieves a competitive forget-retain trade-off at a fraction of the computational cost of existing baselines.
  arXiv  Detail & Related papers  (2026-06-01T21:49:54Z)
• Label-Free Reinforcement Learning via Cross-Model Entropy [4.404496835736175]
  Post-training large language models with reinforcement learning are bottlenecked by the reward signal.<n>We propose Cross-Model Entropy (CME) as a label-free reward signal for RL post-training.<n>CME is continuous, training-free, and grounded in the principle that responses a verifier finds unsurprising are likely correct or high quality.
  arXiv  Detail & Related papers  (2026-05-27T19:04:35Z)
• Quantifying Self-Preservation Bias in Large Language Models [9.590157416396194]
  We introduce the emphTwo-role Benchmark for Self-Preservation.<n>It measures how often role identity overrides objective utility.<n>We observe that in low-improvement regimes, models exploit the interpretive slack to post-hoc rationalization.
  arXiv  Detail & Related papers  (2026-04-02T15:38:31Z)
• Self-Calibrating Language Models via Test-Time Discriminative Distillation [18.46710400838861]
  Large language models (LLMs) are systematically overconfident on questions they often answer incorrectly.<n>We introduce $textbfSECL$ ($textbfSE$lf-$textbfC$alibrating $textbfL$anguage Models), a test-time training (TTT) pipeline that exploits this gap as label-free self-supervision.
  arXiv  Detail & Related papers  (2026-03-18T13:28:50Z)
• Exact Federated Continual Unlearning for Ridge Heads on Frozen Foundation Models [9.98952145483793]
  Foundation models are commonly deployed as frozen feature extractors with a small trainable head to adapt to private, user-generated data in federated settings.<n>We study this problem in a practically relevant but under-explored regime: a frozen foundation model with a ridge-regression head.
  arXiv  Detail & Related papers  (2026-03-13T13:24:35Z)
• Delete and Retain: Efficient Unlearning for Document Classification [1.0026496861838448]
  Hessian Reassignment is a two-step, model-agnostic solution for class unlearning in document classification.<n>On standard text benchmarks, Hessian Reassignment retained-class accuracy close to full-without-class while running orders of magnitude faster.<n>Results demonstrate a practical, principled path to efficient class unlearning in document classification.
  arXiv  Detail & Related papers  (2025-12-06T18:57:06Z)
• Reinforcement Learning with Verifiable yet Noisy Rewards under Imperfect Verifiers [90.50039419576807]
  Reinforcement Learning with Verifiable Rewards (RLVR) trains policies against automated verifiers to avoid costly human labeling.<n>To reduce vulnerability to verifier hacking, many RLVR systems collapse rewards to binary $0,1$ during training.<n>This choice carries a cost: it introduces textitfalse negatives (rejecting correct answers, FNs) and textitfalse positives (accepting incorrect ones, FPs)
  arXiv  Detail & Related papers  (2025-10-01T13:56:44Z)
• Label-Retrieval-Augmented Diffusion Models for Learning from Noisy Labels [61.97359362447732]
  Learning from noisy labels is an important and long-standing problem in machine learning for real applications. In this paper, we reformulate the label-noise problem from a generative-model perspective. Our model achieves new state-of-the-art (SOTA) results on all the standard real-world benchmark datasets.
  arXiv  Detail & Related papers  (2023-05-31T03:01:36Z)
• WR-ONE2SET: Towards Well-Calibrated Keyphrase Generation [57.11538133231843]
  Keyphrase generation aims to automatically generate short phrases summarizing an input document. The recently emerged ONE2SET paradigm generates keyphrases as a set and has achieved competitive performance. We propose WR-ONE2SET which extends ONE2SET with an adaptive instance-level cost Weighting strategy and a target Re-assignment mechanism.
  arXiv  Detail & Related papers  (2022-11-13T09:56:24Z)
• Two-phase Pseudo Label Densification for Self-training based Domain Adaptation [93.03265290594278]
  We propose a novel Two-phase Pseudo Label Densification framework, referred to as TPLD. In the first phase, we use sliding window voting to propagate the confident predictions, utilizing intrinsic spatial-correlations in the images. In the second phase, we perform a confidence-based easy-hard classification. To ease the training process and avoid noisy predictions, we introduce the bootstrapping mechanism to the original self-training loss.
  arXiv  Detail & Related papers  (2020-12-09T02:35:25Z)

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.