FOCUS & RePAIR: Mitigating Text Degeneration via Token-Level Guidance for Pruned Large Language Models

Pruning is a practical approach to compress large language models (LLMs), but it can amplify text degeneration, especially repetition loops, even when perplexity and task accuracy remain largely unchanged. In this work, we present a token-level analysis of this failure mode by viewing decoding as a dynamical process that enters and persists in a small set of recurrent contexts. Our analysis decomposes degeneration into loop entry risk and loop persistence, and shows that persistence is controlled by the escape mass assigned to plausible alternatives within the token sampling set. Motivated by these findings, we propose two token-level guidance objectives for post-pruning fine-tuning. FOCUS reweights distillation toward high-confidence teacher regions to suppress leakage, while RePAIR uses onset-centered positive/negative continuation pairs with a margin loss to promote plausible alternatives and prevent early commitment to repetition loops. Experiments on open-ended continuation and instruction-based generation show that both methods consistently reduce repetition and improve generation quality.

BnBERT-iPET: Sparse Few-Shot Language Modeling for Bengali via Lottery Ticket Pruning

Deep neural networks have shown impressive success in NLP tasks owing to their complex structure and huge number of edges. Achieving state-of-the-art performance in natural language processing with a large pre-trained model such as BERT is expensive and time-consuming, carries a large carbon footprint, and is difficult to realize on machines with minimal computational capability. This creates a barrier to training complex models for resource-constrained languages such as Bengali. However, in a complex neural model, not all edges are equally impactful, and the contributions of some of them can be neglected. Pruning promises to reduce the memory footprint of regular networks, shorten the training time of ever-growing networks, and increase inference efficiency without sacrificing comparable performance. In this work, we introduce BnBERT-iPET, a sparse few-shot language modeling approach for Bengali, and experimentally show that a lightweight few-shot-learned language model retaining only 10% of the edges of an initial model such as BERT can perform neck and neck with much larger models on challenging tasks for a resource-constrained language such as Bengali. By learning from few shots through iterative pattern exploiting training and achieving 90% sparsity with the Lottery Ticket Hypothesis pruning technique, our pruned BnBERT-iPET model proves to be a tough competitor to state-of-the-art language models such as Bangla Electra, Indic-BERT, and XLM-RoBERTa on downstream tasks over standard benchmark datasets of the Bengali language.

Training-Free Knowledge Transfer Across Model Scales through Activation-Guided Pruning

Heterogeneous model fusion seeks to combine models that differ in tasks, initializations, architectures, or scales. We study an underexplored cross-scale setting: improving a small recipient language model with a stronger donor despite substantial architectural mismatch. We ask whether useful capabilities can be transferred without explicit neuron-wise semantic alignment. Building on the observation that truncating a large model to a smaller architecture and injecting it with a tiny mixing weight can already improve the recipient, we propose Activation-Prune-Merge (APM), an activation-guided framework for cross-scale fusion. APM constructs task-conditioned activation maps on the donor, selects salient layers, hidden dimensions, attention heads, and MLP neurons to prune it to the recipient architecture, and injects the resulting donor slice into the original recipient using a micro interpolation coefficient. This formulation treats the donor as a source of concentrated functional components rather than requiring precise structural transplantation. Across 16 benchmarks spanning reasoning, mathematics, code generation, instruction following, and classification, APM improves the overall average accuracy from 55.5% to 60.6% over the original 3B recipient. RTE accuracy increases from 64.3% to 82.3%, QNLI from 52.3% to 65.7%, and BoolQ from 70.8% to 79.2%. Analyses of injection ratios and sequential multi-stage fusion further suggest that activation-guided extraction improves the quality of the transferable donor slice while preserving the small-ratio fusion regime. These results provide evidence that cross-scale heterogeneous fusion can succeed without explicit semantic alignment when the donor contribution is sufficiently concentrated and carefully selected.

Learning What Matters: Supervising Global Context Pruning with Causal Evidence Sets

Pruning a long context means committing to the blocks a model will keep, and the usual selector is distilled from a dense teacher's attention. That assumes attention shows which context the answer depends on. We test the assumption on retrieval tasks where the evidence is known exactly, by masking context and measuring whether the answer changes. Attention and causal dependence disagree. Teachers attend to outdated facts that the answer does not depend on, and they attend differently across training runs that use the same evidence. Selectors trained on that attention copy both failures. On a multi-hop retrieval task, a selector distilled from attention routes at 36% to 98% depending on the training run. The same selector trained on causal evidence sets reaches 99% or better on every run. Dense accuracy does not tell the teachers apart. Masking the frozen teacher recovers the causal sets of these tasks without annotations. Frozen pretrained models show the same conflict, and selectors supervised with known evidence labels beat attention-based eviction through 32B when context must be pruned before the question arrives.

Reasoning Fine-Tuning Induces Persistent Latent Policy States

Reasoning-specialized language models show large performance gains over base models, yet the internal changes responsible for improved multi-step reasoning remain poorly understood. It is unclear whether reasoning fine-tuning improves local token-level competence or globally reorganizes how models structure inference over time. We address this question by modeling Chain-of-Thought reasoning as a switching dynamical system (SDS), in which internal representations evolve under discrete latent policy states. Our framework combines time-aware contrastive representation learning with discrete regime discovery to recover latent policies from activation trajectories. Across four benchmarks and model scales from 1.5B to 32B parameters, reasoning-fine-tuned models exhibit richer latent-policy organization than their base counterparts, characterized by more differentiated transition structure and model-dependent changes in state utilization, persistence, and mixing. The recovered regimes exhibit functional specialization aligned with distinct reasoning stages, and extensive controls confirm that their structure is not explained by correctness, representation learning, or modeling priors, but depends on the coherent temporal organization of reasoning trajectories. Causal interventions further show that the regimes are functionally meaningful: state-swap ablations reduce one-step predictive fit, while transplanting reasoning dynamics into base models improves performance on challenging reasoning problems. Finally, SDS-guided pruning of failure-prone reasoning prefixes outperforms self-consistency in 11 of 12 model-dataset settings, with gains of up to 12.5 percentage points. Together, our results suggest that reasoning fine-tuning globally reorganizes latent dynamics, offering a new lens for mechanistic analysis and process-level control of reasoning models.

Super-Tuning: From Activation-Aware Pruning to Sparse Fine-Tuning

Large language models (LLMs) remain expensive to fine-tune because full-parameter updates require substantial memory, compute, and per-task storage. We study whether saliency signals originally developed for pruning can be reused to choose where a model should adapt. We propose Super, a sparse parameter-efficient fine-tuning (PEFT) method that fixes a small trainable support using a Wanda-style activation-weighted magnitude score [Sun et al., 2023] computed from a calibration pass. We then introduce Supra, a hybrid adapter that combines this sparse update with LoRA while preserving a matched trainable-parameter budget through a simple budget-splitting rule. In single-seed Math17K arithmetic experiments on Llama-3.2-1B and Meta-Llama-3-8B, the best Super/Supra variants achieve the highest average accuracy among the tested schedule-selected adapter configurations. We also include a PaFi-style magnitude-only support as a closest training-free sparse baseline and find that low-score supports under both magnitude and Wanda-style orderings can be effective. These results suggest that simple pruning-inspired orderings can provide useful fixed sparse supports for PEFT, especially when combined with low-rank adapters.

Conditional Co-Ablation: Recovering Self-Repair Backups in Transformer Circuits

Mechanistic interpretability often relies on component-level interventions to discover how a model produces a behavior. This guides attribution, capability knockout, and model pruning downstream to operate by scoring each unit by the effect of ablation in isolation. Such first-order scoring is natural when component importance is additive, but becomes misleading when a transformer self-repairs: after a primary component is removed, a dormant backup can take over, muting the primary's measured effect while the backup itself appears irrelevant on the intact model. We recast this failure as a recovery task, conditional circuit completion, and introduce Conditional Co-Ablation (CoAx), a label-free, output-grounded score that asks how much each remaining unit's ablation effect grows once a primary set has been removed. This conditional growth exposes the second-order interaction that single-unit scores discard. On the GPT-2-small IOI circuit, CoAx raises backup-head recovery from 0.33 to 0.91 ROC-AUC, outperforming all baselines, including self-repair-aware gradient scores (best 0.82); counterfactual patching verifies that the recovered heads causally carry the repair. The same label-free procedure transfers to induction across eight models. Beyond discovery, the recovered backups correct self-repair-masked attribution, identify the components required for capability knockout, and yield repair-aware structured pruning scaling from 124M to 7B. Component importance is therefore not merely an isolated-unit property: in robust circuits, the components that matter can become visible only under the interventions that make them necessary.

SCENIC: Semantic-Conditioned Edge-Aware Neural Framework for Structured IoT Command Generation

Edge Internet of Things (IoT) agents are often constrained by memory capacity, privacy requirements, communication latency, and recurring inference cost. Current smart-home assistants commonly rely on API-level command interfaces or cloud-based language models that remain difficult to deploy on edge devices. This paper addresses edge IoT command generation as a many-to-one structured output task, where multiple natural-language instructions map to the same canonical command string for deterministic smart-home parsing. To support this setting, we propose Semantic-Conditioned Edge-Aware Neural Framework for Structured IoT Command Generation (SCENIC), an end-to-end framework covering model architecture selection, Smart Home Instruct data generation, triplet-loss contrastive supervised fine-tuning, pruning and quantization, and deployment-oriented export. We evaluate sub-0.2B-scale transformer backbones, which are, to the best of our knowledge, among the smallest language-model backbones studied for edge IoT structured command generation. On Smart Home Instruct-Bench, the strongest dense decoder-only row reaches 99.0% EM@1, while the encoder-decoder model retains stronger high-sparsity behavior. A representative pruned INT8 encoder-decoder export preserves 91.0% EM@1 and 99.0% EM@5 while reducing exported model size by 25.38%. TensorRT profiling of the NVIDIA 2:4 sparse encoder export further shows up to 1.8x encoder-component speedup, indicating that the selected encoder-decoder deployment path can retain structured command accuracy under edge-oriented compression while hardware acceleration evidence remains component-level. The SCENIC code and experimental artifacts are open sourced to support reproducibility.

The Benchmark Illusion: Pruned LLMs Can Pass Multiple Choice but Fail to Answer

Compressing large language models reduces memory use and inference cost, but it can also create failures that standard benchmarks miss. A pruned model may still perform well on multiple-choice evaluations, yet fail to answer the same question in open generation. We ask what pruning changes: does it erase the correct answer, or does it make the answer harder to produce as the top output? We study this question with multilingual question answering, tracking the same questions before and after pruning. We find a benchmark illusion. Under high-sparsity pruning, especially Wanda, models often fail in greedy open generation while still selecting the correct answer under multiple-choice scoring. In these recognition-only errors, the answer is usually not gone, but demoted: it often reappears with beam search, sampling, or one in-context example. Overall, multiple-choice benchmarks can overstate the usability of compressed LLMs, creating an evaluation blind spot. Compressed models should be tested on what they can produce, not only on what they can recognize.

Beyond Layer Importance in Layer-wise Sparsity: An Inter-Layer Perturbation-Absorption Perspective

The considerable layer-wise redundancy in large language models (LLMs) has established non-uniform sparsity allocation across layers as the standard pruning approach for efficient compression. Existing layer-wise allocation methods that estimate allocation strategy from local signals such as activation outliers or weight spectra mainly derive from local layer importance, whereas the final post-pruning performance is also influenced by the network's subsequent compensatory capacity. In this paper, we directly characterize this property through controlled perturbation experiments. We make the following empirical findings. First, layers exhibit highly heterogeneous responses to pruning-scale perturbations. In most cases, early layers amplify perturbations, while middle and late layers actively absorb them, with relative L2 drift decreasing monotonically across depth and direction realigning toward the unperturbed hidden-state trajectory. Second, absorption is a large-perturbation phenomenon. Under small perturbations the network exhibits amplification across all layers, and the transition to absorption occurs smoothly as perturbation magnitude grows to pruning scale. This enriches the linearized accumulation theory underlying related works. Building on these findings, we define an absorption coefficient per layer and propose absorption-aware correction, an orthogonal augmentation that improves OWL and AlphaPruning by reducing perplexity by 7.13% and boosting zero-shot accuracy by 1.02% across multiple model families at 70% sparsity.