cs.LGNov 18, 2026

Learning through Discernment: when a small language model learns from user feedback on the job, and when it decays

Authors: Sirisilp Kongsilp

Organizations: Decentralized Science, Inc. · Department of Computer Engineering, Faculty of Engineering, Kasetsart University

Abstract

Small language models on devices could keep improving by learning, overnight, from their own daily work. Without a true outcome, however, something must decide which attempts are worth learning from, and both the model's own judgment and the user's reaction can be wrong. We study a small model (Qwen3 0.6B and 1.7B) with a LoRA adapter, fine-tuned each night on the attempts a judge accepts, over 127 lives of up to a year on WebShop and τ2τ^2-bench. We compare the true outcome, the user's stated verdict and the model's own judgment with a separate, frozen "feedback sense" that reads the user's reaction, built in three versions. Judging itself, the model improved and then decayed, over a year and whatever its adapter. A sense that took polite give-ups for success collapsed 3 of 5 lives; one built for discernment, which first asks whether the user gave a clear verdict, kept 4 of 5 near the oracle for two weeks, then faded on clean but scarce lessons, to the level of self-judging by day 60. When the user cannot see the outcome (τ2τ^2-bench), the sense read reactions correctly, yet 80% of its lessons were failures the customer had thanked. Across 97 lives, the share of failed attempts among the lessons predicted the outcome (Spearman ρρ = -0.81). In our experiments, learning from users held up only where the judge kept failures out and the user could see the outcome; lopsided lessons may also narrow the model.

Figures & tables

Appendix figures & tables5 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

Nov 15, 2026

Logical AI OS: A Training-Free Governance Layer Outside the Model for Instruction Following -- Benefits, Costs, Failures, and the Calibration of the Measuring Instrument Itself

Small language models often fail explicit user requests (length, number of items, format, required or forbidden words, grammatical person, answer language). We measure a training-free governance layer placed outside the model: it reads the requests with rules, checks the answer mechanically, repairs it when a request is not kept, and hands back the original answer when the repair fails. The loop exists in prior work; our contribution is measuring its benefits, costs and failures, and reporting how the measuring instrument itself was calibrated. On four 4-bit Qwen-family models (3B, 7B, 14B, one 7B reasoning model) and 329 Chinese prompts: (1) answers that keep every request the layer read rise significantly (all comparisons p<0.0001), by 34 to 66 of 329 under the corrected checker and 52 to 92 after a length-control fix; a generic redo message brings no significant gain. (2) Exploratory: on 90 repair-triggered prompts, two AI annotators found the layer produced an acceptable answer where the bare model did not more often than the reverse (49:7; 18:12, not significant). (3) Costs: 43-66% more generation time; whole-answer rewriting makes a sizeable share of usable answers unusable. (4) Failures: rule-based reading reaches about 70% and 84% of a two-annotator ceiling on unseen phrasings; unsatisfiable requests make the reasoning model loop; model self-reports are unreliable (18 of 25 "fixed" items were not fixed). (5) Over half of the original checker's "violated" verdicts on structural requests were wrong; after correction the main result holds, and one comparison is downgraded. Conclusions are limited to this Chinese material, the Qwen family and sizes up to 14B. The governance implementation is not released; data, answers, verdicts and the checker are public.
Apr 13, 2026

Disposition Distillation at Small Scale: A Three-Arc Negative Result

We set out to train behavioral dispositions (self-verification, uncertainty acknowledgment, feedback integration) into small language models (0.6B to 2.3B effective parameters) through a four-stage all-MIT distillation pipeline, with follow-on experiments on inference-time attention-head interventions and a frozen-base confidence-gated sidecar. An internal draft reported +33.9-point MCAS and +15.3-point HumanEval gains on a Qwen3-0.6B student; a second-pass sanity check falsified both numbers before publication. The HumanEval delta was a truncation artifact (n_predict=512) that inverted to -8.0 points at n_predict=1024; the MCAS gain disappeared under apples-to-apples scoring. That falsification triggered three subsequent arcs. Across (1) SFT/DPO LoRA on three model families and two domains, (2) inference-time attention-head tempering on o_proj, and (3) a training-free frozen-base sidecar reading the final-token hidden state h_last, we find no operator that moves judge-measured disposition without damaging content or collapsing into stylistic mimicry. The failure is consistent across five models (Qwen3-0.6B, Qwen3-1.7B, Qwen3.5-0.8B, Gemma 4 E2B, and SmolLM2-1.7B-Instruct). A within-distribution cross-validation pass (AUC=0.683) collapsed to chance on fresh prompts (AUC=0.516). We contribute a three-arc negative result with mechanism, a two-failure-mode taxonomy for linear h_last probes, and an honest falsification pipeline that converts the class of false positives we ourselves produced into publishable negatives. As an independent finding, Gemma 4 E2B exhibits near-complete confidence-correctness decoupling on the Chef domain (assertion asymmetry -0.009; the model asserts at 91% regardless of correctness).
Nov 8, 2026

Room to Reconsider: Reply Length Decides Whether Pushback Makes Language Models Sycophantic or Self-Correcting

Language models often abandon a correct answer when a user pushes back without giving a reason, and this behavior is usually studied as sycophancy to be suppressed. The same pushback, however, also leads models to revisit and sometimes repair a wrong answer. We measure both effects on multiple-choice questions from MMLU and ARC-Challenge with four open models from two families (Qwen3 1.7B, 4B and 8B, and OLMo-2 7B), reading the final answer exactly from option logits after a free-form reply. The reply budget changes the verdict. When Qwen3-4B may answer the challenge in at most 96 tokens, pushback leaves accuracy unchanged or slightly lower; with 1,024 tokens, it corrects 43% of its wrong answers and pushback raises accuracy by 5.7 points. Whether pushback helps overall follows a simple break-even rule in the cave and correction rates. Inside the model, caving can be partly predicted before the challenge: hidden states add about 0.04-0.06 AUROC over turn-1 confidence in three of our four probing runs. A probe trained only on self-correction cases predicts caving beyond confidence when replies are truncated, but not when the model can finish its reasoning. Activation steering never reduced caving selectively: where it worked, it suppressed self-correction as well. We conclude that sycophancy evaluations should allow full replies and report self-correction alongside caving.