Library · Machine learning & AI
What Model Pruning & Compression got wrong, from 53 theses
Summary written by a model from the records below: The records document empirical evaluations of neural network compression techniques including weight pruning, quantization, low-rank factorization, and knowledge distillation. Across these studies, compression methods frequently failed when constrained by rigid sparsity patterns, teacher and student capacity mismatches, or operational overheads that eliminated theoretical speedups. These records come from graduate theses, 2021 to 2026, each labeled with its degree where the repository states it, and each links to its thesis. They were extracted by language models reading the full text, so treat each as a lead to read, not a verdict.
Plus 10 problems the authors left open, listed at the end.
Failure modes, by theses reporting them
Select a mode to jump to its records.
Knowledge distillation fails due to teacher capacity gaps, noisy supervision, and representation interference
Large capacity gaps between teacher and student models frequently lead to severe overfitting or ineffective transfer, while noisy teacher logits and overconfidence degrade student guidance. Distillation objectives also suffer from compounded gradient interference across early layers and penalize representation updates during continual learning.
Tried and failed
full model knowledge distillation applied to deep neural network compression. Reason: compounded gradient interference when distilling earlier layers led to diminishing accuracy returns
Everything Is a Matrix: Minimizing Data Movement and Parameter Count Across the Machine Learning Stack · Harvard · PhD
Tried and failed
layerwise hidden representation matching in knowledge distillation applied to cross-architecture language model compression. Outcome: worse than baseline. Reason: unfiltered layerwise intermediate state matching overly constrained student learning compared to output probability distillation
On Parameter Efficiency of Neural Language Models · Georgia Tech · PhD
Tried and failed
knowledge distillation loss with exemplar replay applied to continual learning with recurring classes. Outcome: worse than baseline. Reason: distillation penalizes updating representations when classes reappear across learning sessions with replay
Lifelong Machine Learning with Data Efficiency and Knowledge Retention · EPFL · PhD
Tried and failed
knowledge distillation using sample-dependent eigenbases applied to feature extractor compression. Outcome: did not generalise. Reason: sample-dependent eigenbases produce a sub-optimal representation space that fails to transfer across diverse inputs
Lightweight model for content-style balanced photorealistic style transfer · UT Austin · PhD
Tried and failed
knowledge distillation from large teacher models applied to quantized model compression on small datasets. Outcome: overfit. Reason: Teacher-student capacity mismatch caused severe overfitting on small training datasets
Energy-efficient Neuromorphic Computing for Resource-constrained Internet of Things Devices · Virginia Tech · Doctorate
Tried and failed
knowledge distillation from low-performing teacher models applied to compact neural network training. Outcome: no signal. Reason: student networks cannot acquire useful representations when the teacher model has poor performance
From Symbolic Reasoning to Object Embeddings: Advanced Approaches of Knowledge Distillation in Compacted Neural Networks · Texas Tech · PhD
Tried and failed
distillation using robustly regularized teacher models applied to adversarially robust knowledge distillation. Outcome: worse than baseline. Reason: trade-off regularized teachers provided suboptimal supervision for student accuracy and robustness
The influence of data on adversarial robustness · EPFL · PhD
Tried and failed
fixed loss weighting in knowledge distillation applied to multimodal cross-modal knowledge distillation. Outcome: overfit. Reason: static distillation balance weights caused severe overfitting during knowledge transfer
Effective deep leaning methodologies for salient object detection · Imperial · PhD
Tried and failed
knowledge distillation loss applied to continual learning for image classification. Outcome: worse than baseline. Reason: degraded performance and hindered the ability to benefit from repeated exposures to classes
Shape-biased representations for object category recognition · Georgia Tech · PhD
Tried and failed
knowledge distillation loss for guidance applied to long-term time series forecasting. Outcome: worse than baseline. Reason: distillation loss degraded guidance from the source teacher model over long prediction horizons
Artificial Intelligence for Data-centric Surveillance and Forecasting of Epidemics · Georgia Tech · PhD
Tried and failed
knowledge distillation applied to conditionally gated convolutional neural networks. Outcome: no signal. Reason: small difference between ground truth and teacher output made distilled loss ineffective
Algorithm-Accelerator Co-Design for High-performance and Secure Deep Learning · Cornell · PhD
Tried and failed
knowledge distillation with a large teacher model applied to image classification on complex datasets. Reason: large capacity gap between teacher and student limited effective knowledge transfer
From Symbolic Reasoning to Object Embeddings: Advanced Approaches of Knowledge Distillation in Compacted Neural Networks · Texas Tech · PhD
Tried and failed
knowledge distillation using higher-accuracy teacher model applied to compact object detection student network. Outcome: worse than baseline. Reason: teacher overconfidence degraded distillation quality compared to a smaller teacher model
Improving the Training of Compact Neural Networks for Visual Recognition · EPFL · PhD
Lost to a baseline
On CIFAR-10 distillation with 5000 samples/class, baseline cross-entropy student training reached 91.95%, beating FullGrad + Attention distillation (90.68%).
Gradient-based Methods for Deep Model Interpretability · EPFL · PhD
Lost to a baseline
On CIFAR-10 with standard training under full regularization (dcw), standard training (95.0% accuracy) beat Shrink & Perturb with distillation (94.7% accuracy).
On architectures and training techniques for neural networks · Oxford · PhD
Considered and rejected
Considered and rejected: Rejected direct finetuning of action networks initialized via knowledge distillation without entropy regularization (requiring increased action distribution standard deviation in continuous tasks or explicit low-entropy penalty loss in discrete tasks) due to premature policy entropy collapse.
Using natural language to aid task specification in sequential decision making problems · UT Austin · PhD
Considered and rejected
Considered and rejected: Knowledge distillation: rejected in favor of domain-specific parameter pruning due to difficulties in distilling fine regulatory reasoning into small student models.
Evaluating Efficiency Gains and Security of LLM-Driven Test Generation for Computerised System Validation: A Compliance-Focused Analysis of Life Sciences Testing Processes · Griffith College · Master's
Considered and rejected
Considered and rejected: Rejected traditional logit-based KL divergence distillation because fine-tuned teacher logits were noisy/imprecise and diluted binary classification signals.
Distilled model for contextual soft moderation on social media · Boston University · Master's
Rigid structured and pattern-based pruning degrades model accuracy and destroys critical weights
Enforcing strict filter, channel, or N to M sparsity ratios inadvertently removes essential outlier weights and disrupts complex reasoning capabilities in large language models. These rigid geometric and structural pruning constraints consistently produce inferior task accuracy compared to unconstrained or fine-grained sparsification baselines.
Tried and failed
fixed N:M structured sparsity pruning applied to large language models. Outcome: worse than baseline. Reason: rigid ratio constraints inadvertently remove critical outlier weights, degrading model perplexity
Algorithm–Hardware Co-Design Of Digital Compute-In-Memory Architecture Supporting Flexible And Temporal N:M Sparsity · Georgia Tech · PhD
Tried and failed
structured filter pruning applied to convolutional neural network compression. Outcome: worse than baseline. Reason: degrades model accuracy significantly compared to fine-grained N:M structured sparsity patterns
Scheduling Algorithms for Low-Precision Accumulation on Energy-Efficient Deep Neural Network Hardware · Harvard · PhD
Tried and failed
post-training pruning to structured N:M sparsity applied to large language models. Outcome: worse than baseline. Reason: structured sparsity constraints severely degraded performance on knowledge-intensive downstream tasks
Chasing efficiency in the era of large language models · UT Austin · PhD
Tried and failed
structured weight pruning applied to large language models for reasoning. Outcome: worse than baseline. Reason: removing large fractions of structural weights destroys complex reasoning capabilities and output coherence
Enabling Small Language Models as Efficient and Capable Agents · Virginia Tech · Master's
Tried and failed
structured N:M magnitude-based weight sparsification applied to large language model compression. Outcome: worse than baseline. Reason: magnitude pruning alone severely degrades model perplexity without post-training fine-tuning or decomposition
Structured Sparsity-Aware Hardware-Software Co-Design for Deep Neural Network Acceleration · Georgia Tech · PhD
Tried and failed
activity-based structured state pruning applied to state space sequence models. Outcome: worse than baseline. Reason: aggressive pruning ratios cause non-linear accuracy collapse while latency gains plateau
Performance profiling and activity-based state pruning for efficient Mamba inference · Iowa State · Master's
Lost to a baseline
Structured pruning (e.g., pruning full convolutional filters/channels) loses 1 or more percentage points of accuracy compared to unstructured magnitude pruning at 60% weights remaining on ResNet-50 on ImageNet.
The Lottery Ticket Hypothesis: On Sparse, Trainable Neural Networks · MIT · PhD
Considered and rejected
Considered and rejected: Rejected structured N:M sparsity patterns in favor of unstructured pruning because unstructured pruning performed better and proved more robust.
Odors as ''natural language'': sparse neural networks in mammalian olfactory systems and large language models · Harvard · PhD
Tried and failed
direct computation cost loss penalty applied to dynamic neural network channel pruning. Reason: disproportionately pruned higher-FLOP layers, causing severe layer-wise imbalance compared to threshold loss
Algorithm-Accelerator Co-Design for High-performance and Secure Deep Learning · Cornell · PhD
Tried and failed
Iterative unstructured edge pruning applied to identifying modular reused sub-functions. Reason: Unstructured edge pruning failed to remove dense modular sub-networks without structured unit-level pruning first
Leveraging sparsity in deep neural networks for training efficiency, interpretability and generalization · Georgia Tech · PhD
One-shot and uniform layer pruning degrades representations without adaptive layer budgets or retraining
Allocating uniform pruning budgets across layers ignores non-uniform layer compressibility and causes severe representation damage when pruning early residual blocks or skip connections. Pruning weights in a single shot without iterative retraining or layer-by-layer fine-tuning causes large accuracy drops that subsequent weight regeneration cannot recover.
Tried and failed
post-training magnitude pruning on learned skip-connection weights applied to inter-layer depth-wise averaging modules. Outcome: worse than baseline. Reason: severely degrades model perplexity even at low pruning sparsity thresholds
Enhanced Architectures and Optimization Methods for Efficient Language Modeling · EPFL · PhD
Tried and failed
iterative magnitude pruning without retraining applied to random feature models. Outcome: worse than baseline. Reason: fixed weights without re-optimization degraded representation quality compared to the unpruned minimal L2-norm solution
Adaptive and weighted optimization for efficient and robust learning · UT Austin · PhD
Tried and failed
uniform layer-wise pruning budget allocation applied to deep neural network compression. Outcome: worse than baseline. Reason: fails to account for non-uniform layer compressibility and importance, yielding suboptimal accuracy-compression trade-offs
Efficient Deep Learning: From Theory to Practice · MIT · PhD
Tried and failed
simultaneous pruning of early residual blocks applied to deep neural network compression. Outcome: worse than baseline. Reason: removes critical low-level representations leading to severe performance degradation
Sparsity prior in efficient deep learning based solvers and models · UT Austin · PhD
Tried and failed
backward or parallel layer-wise pruning applied to neural network model compression. Outcome: worse than baseline. Reason: deep-to-shallow or simultaneous pruning degraded accuracy compared to forward layer-by-layer pruning
Overcoming Noise and Variations In Low-Precision Neural Networks · Georgia Tech · PhD
Tried and failed
uninformed magnitude pruning with weight regeneration applied to large language models. Outcome: worse than baseline. Reason: causes irreparable knowledge damage on complex tasks that weight regeneration fine-tuning cannot recover
Chasing efficiency in the era of large language models · UT Austin · PhD
Tried and failed
iterative magnitude pruning on random feature models applied to learning functions with dense representations. Reason: dense target representations dilute the performance benefits of magnitude pruning
Adaptive and weighted optimization for efficient and robust learning · UT Austin · PhD
Tried and failed
post-training channel pruning without fine-tuning applied to convolutional neural networks. Outcome: worse than baseline. Reason: severe accuracy degradation exceeding ten percentage points with negligible throughput improvement under hardware execution
Efficient AI model acceleration through tensor decomposition and dataflow architecture design · Imperial · PhD
Considered and rejected
Considered and rejected: Rejected full-network pruning in favor of dense-only layer pruning to avoid disrupting critical pre-trained convolutional feature filters
MODEL OPTIMIZATION FOR VEHICLE RECOGNITION ON EDGE DEVICES · Naval Postgraduate School · Master's
Considered and rejected
Considered and rejected: Rejected one-shot magnitude pruning in favor of iterative magnitude pruning because gradual pruning yields higher accuracy and sparser matching subnetworks.
The Lottery Ticket Hypothesis: On Sparse, Trainable Neural Networks · MIT · PhD
Tried and failed
Pruning feedforward versus recurrent connections applied to spiking neural networks. Outcome: worse than baseline. Reason: feedforward sparsity significantly degraded phoneme error rate compared to recurrent pruning
Biologically Inspired Spiking Neural Networks for Speech Recognition · EPFL · PhD
Sparse models fail to deliver practical latency reductions and hardware throughput improvements
Unstructured sparse weights and pruned network structures often require specialized sparse accelerators and structural metadata, resulting in negligible throughput gains or even lower inference framerates than dense baselines. In other settings, algorithmic compression and pruning routines introduce substantial computation or metadata overhead that outweighs the memory and bandwidth benefits.
Considered and rejected
Considered and rejected: Rejected unstructured pruning for learning compact hidden representations because pruned sparse weights do not reduce the number of features or inference overhead.
Achieving More with Less: Learning Generalizable Neural Networks With Less Labeled Data and Computational Overheads · Virginia Tech · Doctorate
Considered and rejected
Considered and rejected: Unstructured network pruning: rejected in favor of structured pruning to avoid requiring specialized sparse hardware accelerators and transmission of sparse structural metadata.
Deep joint source-channel coding for vision-based inference at the wireless edge · Imperial · PhD
Tried and failed
run-length encoding metadata compression applied to sparse tensor accelerator bandwidth optimization. Reason: bandwidth overhead is dominated by data payload rather than metadata encoding size
Systematic Modeling and Design of Sparse Deep Neural Network Accelerators · MIT · PhD
Tried and failed
weight pruning applied to bound propagation neural network verification. Outcome: too slow. Reason: Sparsity does not yield computational speedups in symbolic interval propagation routines despite reducing constraint complexity in SMT/MILP
Robust neural networks: verification, training, and repair · Imperial · PhD
Tried and failed
decision tree pruning via node importance filtering applied to multi-layer decision trees. Outcome: too slow. Reason: filtering individual node weights did not reduce the number of sampling layers required for computation
A Methodology for Forecasting and Mitigating Risk in Cislunar Mission Planning Using Real Options · Georgia Tech · PhD
Lost to a baseline
Embedded-ViT with 50% pruning at 128x128 input resolution on CPU achieved lower throughput (23.36 FPS) than unpruned DAE-Former baseline (31.32 FPS)
Weakly supervised and embedded semantic segmentation for Computer Aided Diagnostics · TU Wien · PhD
Lost to a baseline
The baseline accelerator from [29] is ×1.85 faster than the proposed bit-level pruning design due to its dataflow latency optimization.
Design of Efficient DNN Accelerator Architectures · Saskatchewan · PhD
Considered and rejected
Considered and rejected: Rejected dynamic pruning during local training because resource-constrained edge clients cannot train dense unpruned DNNs initially.
REFT: Resource-Efficient Federated Training Framework for Heterogeneous and Resource-Constrained Environments · Virginia Tech · Master's
Considered and rejected
Considered and rejected: Rejected pseudo-rendering and multi-view importance ranking for 3DGS pruning due to expensive on-the-fly summation and GPU compute overhead.
LOW PRECISION AND SPARSE AI ALGORITHMS FOR EDGE COMPUTING · Cornell · PhD
Considered and rejected
Considered and rejected: Rejected alternating iterative SVD updates and outlier extraction during KV cache compression due to severe latency overheads unacceptable for generative inference.
On the Efficiency and Steerability of Self-Attention Mechanism of Large Language Models · Georgia Tech · PhD
Early pruning and initialization pruning methods underperform random sparsity and post-training baselines
Intelligent pruning algorithms applied at initialization or early in training frequently fall behind simple random pruning baselines or suffer catastrophic layer collapse. Post-training magnitude pruning and gradual pruning methods consistently outperform initialization-time techniques across high-sparsity regimes.
Lost to a baseline
Magnitude pruning after training outperformed all initialization pruning methods including PHEW across high-sparsity regimes.
Leveraging sparsity in deep neural networks for training efficiency, interpretability and generalization · Georgia Tech · PhD
Lost to a baseline
Global random pruning and random reinitialization match or beat IMP at step 0 on standard ResNet-20 (88.6% and 88.8% vs 88.5% test accuracy at 16.8% density).
The Lottery Ticket Hypothesis: On Sparse, Trainable Neural Networks · MIT · PhD
Lost to a baseline
Repeated full pruning and gradual pruning were beaten by simple random sparse initialization on large MNIST networks (2-3 hidden layers, 200 neurons/layer)
Overcoming Noise and Variations In Low-Precision Neural Networks · Georgia Tech · PhD
Lost to a baseline
Stat channel pruning on Resnet-20 performed worse than random pruning across most preservation rates.
Fine Granularity is Critical for Intelligent Neural Network Pruning · York University · Master's
Lost to a baseline
At very low channel preservation rates on Resnet-20, intelligent pruning methods (SNIP, SNat, Stat) were outperformed by random pruning at initialization.
Fine Granularity is Critical for Intelligent Neural Network Pruning · York University · Master's
Lost to a baseline
On DBSN node pruning at low preservation rates, intelligent pruning methods dropped below random pruning due to over-pruning the final bottleneck layer.
Fine Granularity is Critical for Intelligent Neural Network Pruning · York University · Master's
Lost to a baseline
On Tiny-ImageNet, Initial (Weight) Magnitude Pruning suffered catastrophic accuracy drops due to layer collapse at lower network densities compared to PHEW, SynFlow, and SynFlow-L2.
Leveraging sparsity in deep neural networks for training efficiency, interpretability and generalization · Georgia Tech · PhD
Lost to a baseline
One-time pruning underperformed random sparsity initialization on small neural networks across multiple wait times
Overcoming Noise and Variations In Low-Precision Neural Networks · Georgia Tech · PhD
Lost to a baseline
On CIFAR-10 ResNet-50 pruning, 1-epoch early magnitude pruning performed worse than SNIP-MB and gradual pruning methods.
Towards efficient and reliable neural networks · Oxford · PhD
Aggressive low-bit quantization and indiscriminate low-rank factorization cause severe representation errors
Extreme sub-four-bit quantization and post-training quantization without distillation introduce severe rounding errors that disrupt weight orders and degrade accuracy. Applying uniform low-rank matrix approximation or tensor decomposition across all layers harms critical lower-level representations compared to targeted layer-specific pruning.
Tried and failed
post-training quantization applied to large language models across task difficulties. Outcome: did not generalise. Reason: performance degradation was non-monotonic across varying task difficulty levels compared to pruning
Chasing efficiency in the era of large language models · UT Austin · PhD
Tried and failed
quantization before magnitude pruning applied to neural network model compression. Outcome: worse than baseline. Reason: quantization perturbs weight magnitudes, disrupting pruning order and degrading perplexity
Compressing DNNs Using Microscaling Formats with Sensitivity and Sparsity · EPFL · PhD
Tried and failed
post-training quantization without knowledge distillation applied to deep neural networks. Outcome: worse than baseline. Reason: aggressive low-bit term quantization severely degrades accuracy without multi-resolution distillation guidance
Systolic Architectures for Efficient Deep Neural Network Implementations with Assured Performance · Harvard · PhD
Tried and failed
extreme sub-4-bit iterative weight quantization hierarchy applied to deep neural network weight compression. Outcome: worse than baseline. Reason: drastic rounding errors at 1-bit and 2-bit precisions severely degrade network performance
Advancing efficiency and trustworthiness : from computer vision to multimodal large language models · UT Austin · PhD
Tried and failed
uniform low-rank matrix approximation across all layers applied to deep neural network weight matrices. Outcome: worse than baseline. Reason: indiscriminate rank reduction harms critical representations compared to targeted layer-specific pruning
Discovering and Engineering the Computation Underlying Large Intelligent Agents · MIT · PhD
Considered and rejected
Considered and rejected: Rejected post-training integer quantization (INT8/INT4) in favor of FP16/FP32 encoder pruning to avoid accuracy degradation and retraining/calibration overhead.
On-device Learning and Inference Optimization for Lightweight Neural Networks and Transformers on Microcontrollers · Politecnico di Torino · PhD
Lost to a baseline
Without knowledge distillation, SVD-only compressed ResNet-50 drops to 5.73% accuracy on Tiny-ImageNet at rank 32.
Everything Is a Matrix: Minimizing Data Movement and Parameter Count Across the Machine Learning Stack · Harvard · PhD
Tried and failed
tensor-train decomposition for neural network compression applied to dense language model layers. Outcome: worse than baseline. Reason: degrades broad, weak lower-level feature representation required for commonsense reasoning tasks
Generative language model compression with algebraic approaches · Imperial · PhD
Considered and rejected
Considered and rejected: Rejected conventional bit packing because the wide dynamic range of ML values provides minimal compression across individual tensors.
Mitigating the Impact of Data Movement in Memory-Intensive Applications · University of Washington · PhD
Left open by the authors
Problems the authors named and did not get to.
Left open
Analyze why BERT pruning algorithms naturally prune more neurons in deeper layers than earlier layers. Blocker: None
Sparsity prior in efficient deep learning based solvers and models · UT Austin · PhD
Left open
Develop model compression techniques using highly sparse binary masks to reduce memory footprint and inference latency without degrading accuracy. Blocker: None
Algorithms for Efficient and Robust Distributed Deep Learning · EPFL · PhD
Left open
Develop pruning techniques to recover accuracy for deep neural network models with dual-side structured sparsity (HighLight_DSSO). Blocker: Lacks specific targets, pruning algorithms, or designated network architectures to evaluate
Systematic Modeling and Design of Sparse Deep Neural Network Accelerators · MIT · PhD
Left open
Tune training hyperparameters to quantify the workload required to recover accuracy after deep learning model pruning. Blocker: None
Tools for efficient Deep Learning · Imperial · PhD
Left open
Apply pruning, feature importance, and complexity analysis methods to CNN models representing binary nanophotonic structures. Blocker: None
Machine Learning Approaches for Knowledge Discovery in Nanophotonic Structures · Georgia Tech · PhD
Left open
Implement layer-wise pruning on 4D CNNs to evaluate temporal feature retention and parameter reduction in deep layers. Blocker: None
Development of a machine-learning platform for the autonomous analysis of 3D+T calcium imaging data · UT Austin · PhD
Left open
Extend the pruning and feature importance framework to CNN models trained on binary nanophotonic structure electromagnetic simulation data. Blocker: None
Machine Learning Approaches for Knowledge Discovery in Nanophotonic Structures · Georgia Tech · PhD
Left open
Apply advanced neural network pruning to reduce connection complexity between design and response spaces in the nanophotonic inverse design framework. Blocker: None
A New Paradigm for Knowledge Discovery and Design in Nanophotonics Based on Artificial Intelligence · Georgia Tech · PhD
Left open
Apply pruning, quantization, and edge hardware acceleration to the STFT-CNN power quality disturbance classification model. Blocker: None
Power Quality Disturbance Classification in IEEE 9-Bus System using STFT and Deep Learning · Texas Tech · Master's
Left open
Investigate why structured pruning underperforms random initialization on larger networks trained with analog noise on MNIST. Blocker: None
Overcoming Noise and Variations In Low-Precision Neural Networks · Georgia Tech · PhD
Who has worked on this
Authors of the public theses cited on this page, most-cited first.
- Devon D. Janke · Georgia Tech · PhD · 2021
Overcoming Noise and Variations In Low-Precision Neural Networks - Shreyas Malakarjun Patil · Georgia Tech · PhD · 2024
Leveraging sparsity in deep neural networks for training efficiency, interpretability and generalization - Jonathan Frankle · MIT · PhD · 2023
The Lottery Ticket Hypothesis: On Sparse, Trainable Neural Networks - Andrew Baldwin Heyman · York University · Master's · 2023
Fine Granularity is Critical for Intelligent Neural Network Pruning - Andrew Sabot · Harvard · PhD · 2025
Everything Is a Matrix: Minimizing Data Movement and Parameter Count Across the Machine Learning Stack - Himel Das Gupta · Texas Tech · PhD · 2025
From Symbolic Reasoning to Object Embeddings: Advanced Approaches of Knowledge Distillation in Compacted Neural Networks - Yannan Wu · MIT · PhD · 2023
Systematic Modeling and Design of Sparse Deep Neural Network Accelerators - Weizhe Hua · Cornell · PhD · 2022
Algorithm-Accelerator Co-Design for High-performance and Secure Deep Learning - Akshat Ramachandran · Georgia Tech · PhD · 2026
Algorithm–Hardware Co-Design Of Digital Compute-In-Memory Architecture Supporting Flexible And Temporal N:M Sparsity - Yuhan Chen · Boston University · Master's · 2026
Distilled model for contextual soft moderation on social media - Gaurav Srivastava · Virginia Tech · Master's · 2026
Enabling Small Language Models as Efficient and Capable Agents - Abdullah Al Asif · Iowa State · Master's · 2026
Performance profiling and activity-based state pruning for efficient Mamba inference
Checking a claim in this area?
We can run the same search on any method or claim. If nothing turns up, we will say so, and that proves nothing on its own.