Bottlenecks

Follow this field Ask about this field

Recent momentum

emerging

0 papers in the last 28 days · 0.0% of indexed attention

Twelve weeks of publication activity for this topic as it is defined today.

Weekly history

Recent digests

What was published in this field, kept on the site without email delivery.

Period ending 2026-09-21

3 new papers

A weekly snapshot of new work published in Bottlenecks.

Period ending 2026-09-14

5 new papers

A weekly snapshot of new work published in Bottlenecks.

Inside this field

Focused directions

Focused topic

Bottlenecks

Within Bottlenecks

175 papersExplore

Focused topic

Concept Bottleneck Models

Within Bottlenecks

63 papersExplore

Focused topic

Class-Incremental Concept Bottleneck Model

Within Bottlenecks

19 papersExplore

227 papers

Latest in Bottlenecks

Browse all topics →

Architecture, Not Scale: Circuit Localization in Large Language Models
May 9, 2026Sohan VenkateshMechanistic Interpretability Circuits
SynerDiff: Synergetic Continuous Batching for Fast and Parallel Diffusion Model Inference
May 9, 2026Ziqi Zhou, Peng Yang, Yuxin Liang +2Well-Tuned Static Batching Policy Optimal Scheduling
Too Many Specialists: Emergent Inefficiencies and Bottlenecks for Multi-agent Ad-hoc Collaboration
May 8, 2026Benjamin Panny, Shashank Mehrotra, Zahra Zahedi +2Collaboration Bottlenecks
ShifaMind: A Multiplicative Concept Bottleneck for Interpretable ICD-10 Coding
May 8, 2026Mohammed Sameer Syed, Xuan LuInternational Classification Of Disease Codes Concept Bottleneck Models
Decoupling Endpoint and Semantic Transition Learning for Zero-Shot Composed Image Retrieval
May 8, 2026Mingyu Liu, Sihan Huang, Yijia Fan +4Composed Image Retrieval Single Images
Radiologist-Guided Causal Concept Bottleneck Models for Chest X-Ray Interpretation
May 8, 2026Amy Rafferty, Rishi Ramaesh, Ajitha RajanChest X-Ray Class-Incremental Concept Bottleneck Model
What If We Let Forecasting Forget? A Sparse Bottleneck for Cross-Variable Dependencies
May 8, 2026Fan Zhang, Shiming Fan, Hua WangMultivariate Time Series Forecasting Multivariate Forecasting
Hyperbolic Concept Bottleneck Models
May 7, 2026Daniel Uyterlinde, Swasti Shreya Mishra, Pascal MettesClass-Incremental Concept Bottleneck Model Concept Bottleneck Models
OBLIQ-Bench: Exposing Overlooked Bottlenecks in Modern Retrievers with Latent and Implicit Queries
May 7, 2026Diane Tchuindjo, Devavrat Shah, Omar KhattabRetrievers Search Algorithms
Safety Anchor: Defending Harmful Fine-tuning via Geometric Bottlenecks
May 7, 2026Guoxin Lu, Letian Sha, Qing Wang +4Harmful Fine-Tuning Large Language Model Safety
TinySSL: Distilled Self-Supervised Pretraining for Sub-Megabyte MCU Models
May 7, 2026Bibin WilsonSelf-Supervised Chain-Of-Thought Distillation
On the (In-)Security of the Shuffling Defense in the Transformer Secure Inference
May 6, 2026Zhengyi Li, Yakai Wang, Kang Yang +6Inference Engine Security
Temporal Reasoning Is Not the Bottleneck: A Probabilistic Inconsistency Framework for Neuro-Symbolic QA
May 5, 2026Tran Quang LiemNeuro-Symbolic Framework Reasoning Failures
MEMTIER: Tiered Memory Architecture and Retrieval Bottleneck Analysis for Long-Running Autonomous AI Agents
May 5, 2026Bronislav Sidik, Lior RokachAgent Memory Episodic Experiences
The Productivity-Reliability Paradox: Specification-Driven Governance for AI-Augmented Software Development
May 1, 2026Sabry E. FarragSoftware Engineering Ai-Ran
Scaling Federated Linear Contextual Bandits via Sketching
May 1, 2026Hantao Yang, Hong Xie, Xutong Liu +1Linear Contextual Bandits Federated Learning
Beyond Gaussian Bottlenecks: Topologically Aligned Encoding of Vision-Transformer Feature Spaces
Apr 30, 2026Andrew Bond, Ilkin Umut Melanlioglu, Erkut Erdem +13D Encoder Variational Autoencoder
When Continual Learning Moves to Memory: A Study of Experience Reuse in LLM Agents
Apr 29, 2026Qisheng Hu, Quanyu Long, Wenya WangContinual Learning Methods Large Language Model Agents
Looking for the Bottleneck in Fine-grained Temporal Relation Classification
Apr 27, 2026Hugo Sousa, Ricardo Campos, Alípio JorgeRelation Extraction Relational Deep Learning
Credal Concept Bottleneck Models for Epistemic-Aleatoric Uncertainty Decomposition
Apr 27, 2026Tanmoy Mukherjee, Thomas Bailleux, Pierre Marquis +1Epistemic Uncertainty Class-Incremental Concept Bottleneck Model
Enhanced Privacy and Communication Efficiency in Non-IID Federated Learning with Adaptive Quantization and Differential Privacy
Apr 25, 2026Emre Ardıç, Yakup GençFederated Learning Differential Privacy
REC-CBM: Rubric-Aware Error-Correction Concept Bottleneck Models for Trustworthy Open-Ended Grading
Apr 24, 2026Chengshuai Zhao, Fan Zhang, Kumar Satvik Chaudhary +4Grading Concept Bottleneck Models
Symbolic Grounding Reveals Representational Bottlenecks in Abstract Visual Reasoning
Apr 23, 2026Mohit Vaishnav, Tanel TammetVisual Reasoning Large Vision Language Models
Concept Inconsistency in Dermoscopic Concept Bottleneck Models: A Rough-Set Analysis of the Derm7pt Dataset
Apr 21, 2026Gonzalo Nápoles, Isel Grau, Yamisleydi SalgueiroClass-Incremental Concept Bottleneck Model Skin Lesion Classification
AQPIM: Breaking the PIM Capacity Wall for LLMs with In-Memory Activation Quantization
Apr 20, 2026Kosuke Matsushima, Yasuyuki Okoshi, Masato Motomura +1Graphics Processing Unit Memory LLM Inference Efficiency
Heterogeneity in Formal Linguistic Competence of Language Models: Is Data the Real Bottleneck?
Apr 20, 2026H S V N S Kowndinya Renduchintala, Sumit BhatiaLarge Language Model Performance Linguistics
Predictive Multi-Tier Memory Management for KV Cache in Large-Scale GPU Inference
Apr 19, 2026Sanjeev Rao GanjihalKey-Value Cache Graphics Processing Unit Memory
Capacity-Controlled Global Attention for Graph Transformers
Apr 19, 2026Yang Liu, Dongxin Guo, Tom Zheng +3Heterogeneous Graph Transformers Transformer Attention
Tree of Concepts: Interpretable Continual Learners in Non-Stationary Clinical Domains
Apr 18, 2026Dongkyu Cho, Xiyue Li, Samrachana Adhikari +1Efficient Continual Learning Decision Trees
KAIROS: Stateful, Context-Aware Power-Efficient Agentic Inference Serving
Apr 17, 2026Yichao Yuan, Mosharaf Chowdhury, Nishil TalatiBottlenecks
Complete Cyclic Subtask Graphs for Tool-Using LLM Agents: Flexibility, Cost, and Bottlenecks in Multi-Agent Workflows
Apr 17, 2026Luay Gharzeddine, Samer SaabMulti-Agent Workflows Multi-Turn Tool-Use Agent
Prototype-Grounded Concept Models for Verifiable Concept Alignment
Apr 17, 2026Stefano Colamonaco, David Debot, Pietro Barbiero +1Concept Bottleneck Models Model Interpretability
Placing Puzzle Pieces Where They Matter: A Question Augmentation Framework for Reinforcement Learning
Apr 17, 2026Yangyi Fang, Jiaye Lin, Xiaoliang Fu +2Puzzle Augmentation
Concept-wise Attention for Fine-grained Concept Bottleneck Models
Apr 17, 2026Minghong Zhong, Guoshuai Zou, Kanghao Chen +2Fine-Grained Perception Concept Bottleneck Models
Exploiting Correlations in Federated Learning: Opportunities and Practical Limitations
Apr 16, 2026Adrian Edin, Michel Kieffer, Mikael Johansson +1Federated Learning Spurious Correlations
CI-CBM: Class-Incremental Concept Bottleneck Model for Interpretable Continual Learning
Apr 16, 2026Amirhosein Javadi, Tuomas Oikarinen, Tara Javidi +1Efficient Continual Learning Continual Learning Methods
MAPLE: Metadata Augmented Private Language Evolution
Feb 26, 2026Eli Chien, Yuzheng Hu, Ryan McKenna +3Synthetic Data Alphaevolve
CauScale: Neural Causal Discovery at Scale
Feb 9, 2026Bo Peng, Sirui Chen, Jiaguo Tian +2Causal Foundation Models Causal Reasoning
From One-to-One to Many-to-Many: Dynamic Cross-Layer Injection for Deep Vision-Language Fusion
Jan 15, 2026Cheng Chen, Yuyu Guo, Pengpeng Zeng +4Large Vision Language Models Vision Encoders
Identifying and Mitigating Bottlenecks in Role-Playing Agents: A Systematic Study of Disentangling Character Profile Axes
Jan 8, 2026Yonghyun Jun, Junhyuk Choi, Jeonghyun Park +3Role-Playing Agents Characterization
A Tool Bottleneck Framework for Clinically-Informed and Interpretable Medical Image Understanding
Dec 24, 2025Christina Liu, Alan Q. Wang, Joy Hsu +2Medical Vision-Language Models Medical Image Classification
The Inductive Bottleneck: Data-Driven Emergence of Representational Sparsity in Vision Transformers
Dec 8, 2025Kanishk AwadhiyaVision Transformer Imagenet
Parameter-Efficient Construction of the Rashomon Slice for Concept Bottleneck Models
Nov 24, 2025Shihan Feng, Cheng Zhang, Michael Xi +3Concept Bottleneck Models Model Selection
Beyond Reactivity: Measuring Proactive Problem Solving in LLM Agents
Oct 22, 2025Gil Pasternak, Dheeraj Rajagopal, Julia White +4Proactive Large Language Model Agents
Subspace Networks: Scaling Decentralized Training with Communication-Efficient Model Parallelism
Jun 2, 2025Sameera Ramasinghe, Thalaiyasingam Ajanthan, Gil Avraham +2Large Language Model Compression Billion-Parameter Models
VIBE: Annotation-Free Video-to-Text Information Bottleneck Evaluation for TL;DR
May 23, 2025Shenghui Chen, Po-han Li, Sandeep Chinchali +1Video Captioning Summarization
Quantum Adaptive Self-Attention for Quantum Transformer Models
Apr 5, 2025Chi-Sheng Chen, En-Jui KuoQuantum Machine Learning Quantum Advantage

May 9, 2026 · cs.CLJ/K move · Enter open · S save

Architecture, Not Scale: Circuit Localization in Large Language Models

Sohan Venkatesh

Manipal Institute of Technology Bengaluru.

Mechanistic interpretability assumes that circuit analysis becomes harder as models scale. We challenge this assumption by showing that the attention architecture matters more than parameter count. Studying three circuit types across Pythia and Qwen2.5, we find that grouped query attention produces circuits that are far more concentrated and mechanistically stable than standard multi-head attention at comparable scales. The same concentration pattern holds across indirect object identification, induction heads, and factual recall. Within a single architecture family (Qwen2.5), factual recall circuits undergo a discrete phase transition above a critical scale, collapsing to a single bottleneck rather than degrading gradually. These findings suggest that some architectural choices make large models more tractable to study and that interpretability difficulty is not a fixed consequence of model size.

Mechanistic Interpretability Circuits Concept Bottleneck Models Concentration Factual Recall

Open paper Ask about it PDF