ACL 2026 — Code & Software Engineering | Rajat Kanti Bhattacharjee

Code & Software Engineering

100 papers Links not yet available — ACL proceedings pending on ACL Anthology.

Sense and Sensitivity: Examining the Influence of Semantic Recall on Long Context Code Understanding
CRISP: Persistent Concept Unlearning via Sparse Autoencoders
What is a protest anyway? Codebook conceptualization is still a first-order concern in LLM-era classification
Mechanistic Interpretability Should Prioritize Feature Consistency in Sparse Autoencoders
CodeHacker: Automated Test Case Generation for Detecting Vulnerabilities in Competitive Programming Solutions
SAC: Neural Speech Codec with Semantic-Acoustic Dual-Stream Quantization
TRACE: Evaluating Execution Efficiency of LLM-Based Code Translation
Bridging Language and Items for Retrieval and Recommendation: Benchmarking LLMs as Semantic Encoders
CODERL+: Improving Code Generation via Reinforcement with Execution Semantics Alignment
Saber: Efficient Sampling with Adaptive Acceleration and Backtracking Enhanced Remasking for Diffusion Language Model in Code Generation
CodeFlowBench: A Multi-turn, Iterative Benchmark for Complex Code Generation
VideoPro: Adaptive Program Reasoning for Long Video Understanding
ChartVerse: Scaling Chart Reasoning via Reliable Programmatic Synthesis from Scratch
From Completion to Editing: Unlocking Context-Aware Code Infilling via Search-and-Replace Instruction Tuning
Beyond Monolingual Assumptions: A Survey on Code-Switched NLP in the Era of Large Language Models across Modalities
MulVul: Retrieval-augmented Multi-Agent Code Vulnerability Detection via Cross-Model Prompt Evolution
From Laboratory to Real-World Applications: Benchmarking Agentic Code Reasoning at the Repository Level
XY-Tokenizer: Mitigating the Semantic-Acoustic Conflict in Low-Bitrate Speech Codecs
“Newspaper Eat” Means “Not Tasty”: A Taxonomy and Benchmark for Coded Language in Real-World Chinese Online Reviews
CodeEvo: Interaction-Driven Synthesis of Code-centric Data through Hybrid and Iterative Feedback
CoQuIR: A Comprehensive Benchmark for Code Quality-Aware Information Retrieval
Lifting Optimized Binaries to Canonical Compiler IR via Structure-Aware Retrieval and Iterative Verification
WebCoderBench: Benchmarking Web Application Generation with Comprehensive and Interpretable Evaluation Metrics
Adaptive Prompt Structure Factorization: A Framework for Self-Discovering and Optimizing Compositional Prompt Programs
Cognitive Alpha Mining via LLM-Driven Code-Based Evolution
KV-Embedding: Training-free Text Embedding via Internal KV Re-routing in Decoder-only LLMs
CODESTRUCT: Code Agents over Structured Action Spaces
From Charts to Code: A Hierarchical Benchmark for Multimodal Models
ExecVerify: White-Box RL with Verifiable Stepwise Rewards for Code Execution Reasoning
Map of Encoders – Mapping Sentence Encoders using Quantum Relative Entropy
ReFEree: Reference-Free and Fine-Grained Method for Evaluating Factual Consistency in Real-World Code Summarization
When “Correct” Is Not Safe: Can We Trust Functionally Correct Patches Generated by Code Agents?
ParaCodex: A Profiling-Guided Autonomous Coding Agent for Reliable Parallel Code Generation and Translation
The Path Not Taken: Duality in Reasoning about Program Execution
SpecAgent: A Speculative Retrieval and Forecasting Agent for Code Completion
Taming System Complexity: Demystifying Software Engineering Agents in Diagnosing Linux Kernel Faults
DisCo_Speech: Controllable Zero-Shot Speech Generation with A Disentangled Speech Codec
Hierarchical Token Prepending: Enhancing Information Flow in Decoder-based LLM Embeddings
StoryCoder: Narrative Reformulation for Structured Reasoning in LLM Code Generation
CodeJudgeBench: Benchmarking LLM-as-a-Judge for Coding Tasks
DeepGuard: Secure Code Generation via Multi-Layer Semantic Aggregation
Benchmarking and Enabling Efficient Chinese Medical Retrieval via Asymmetric Encoders
HSCodeComp: A Realistic and Expert-level Agent Benchmark for Hierarchical Rule Application
Aligned Multi-View Scripts for Universal Chart-to-Code Generation
Language on Demand, Knowledge at Core: Composing LLMs with Encoder-Decoder Translation Models for Extensible Multilinguality
Controllable LLM Reasoning via Sparse Autoencoder‑Based Steering
Causal2Vec: Improving Decoder-only LLMs as Embedding Models through a Contextual Token
FormalScience: Scalable Human-in-the-Loop Autoformalisation of Science with Agentic Code Generation in Lean
BTC-LLM: Efficient Sub-1-Bit LLM Quantization via Learnable Transformation and Binary Codebook
Scaling Laws for Code: A More Data-Hungry Regime
CharTide: Data-Centric Chart-to-Code Generation via Tri-Perspective Tuning and Inquiry-Driven Evolution
EyeMulator: Improving Code Language Models by Mimicking Human Visual Attention
AIRCoder: Adaptive Integration of Multi-dimensional Retrieval for Repository-level Code Completion
From Logical to Computational Sparsity: Structure-Aware Block-Sparse Attention for Long-Code Completion
Selective Knowledge Distillation: Fusing LLM Semantic Strengths with DNN Efficiency for Binary Code Similarity Detection
CoreCodeBench: Decoupling Code Intelligence via Fine-Grained Repository-Level Tasks
Compiling Activation Steering into Weights via Null-Space Constraints for Stealthy Backdoors
KoCo-Bench: Can Large Language Models Leverage Domain Knowledge in Software Development?
QiMeng-PRepair: Precise Code Repair via Edit-Aware Reward Optimization
EGSS: Entropy-guided Stepwise Scaling for Reliable Software Engineering
MavenCoder: Competitive Code Generation via Model Adaptive Planning Strategies and Multi-Perspective Verification Enhancement
Do LLMs Encode Functional Importance of Reasoning Tokens ?
UniVocal: Unified Speech-Singing Code-Switching Synthesis
RedCoder: Automated Multi-Turn Red Teaming for Code LLMs
MARS²: Scaling Multi-Agent Tree Search via Reinforcement Learning for Code Generation
Multilingual Language Models Encode Script Over Linguistic Structure
Beyond Code Pairs: Dialogue-Based Data Generation for LLM Code Translation
MMSciCode: Real-world Evaluation of Multilingual Multi-Discipline Scientific Research Coding
Corpus-Dependent Subcharacter Encoding via HMM-Guided Code Assignment
E2EDev: Benchmarking Large Language Models in End-to-End Software Development Task
Gated Tree Cross-Attention for Checkpoint-Compatible Syntax Injection in Decoder-Only LLMs
SeDev: Structured Semantic Exploration for LLM-Driven Code Generation
EVM-QuestBench: An Execution-Grounded Benchmark for Natural-Language Transaction Code Generation
SLICEFORMER: Static Program Slicing Using Language Models With Dataflow-Aware Pretraining and Constrained Decoding
TexOCR: Advancing Document OCR Models for Compilable Page-to-LaTeX Reconstruction
LeCoDe: A Benchmark Dataset for Interactive Legal Consultation Dialogue Evaluation
Bootstrapping Code Translation with Weighted Multilanguage Exploration
PaT: Planning-after-Trial for Efficient Test-Time Code Generation
Frozen LLMs are Native Decoders for High-Norm Semantic Vectors
Example Quality Matters: Multi-Aspects Example Augmentation for Private Library Programming
Dashboard2Code: Evaluating Multimodal Models on Reconstructing Interactive Dashboards
Omni-I2C: A Holistic Benchmark for High-Fidelity Image-to-Code Generation
CodeRipple: Wavelet-Based Detection of LLM-Generated Code
SciCoQA: Quality Assurance for Scientific Paper–Code Alignment
From Where Words Come: Efficient Regularization of Code Tokenizers Through Source Attribution
PDTrim: Targeted Pruning for Prefill-Decode Disaggregation in Inference
RealChart2Code: Bridging the Gap in Real-World Chart-to-Code Generation via Multi-Task Evaluation
From Proof to Program: Characterizing Tool-Induced Reasoning Hallucinations in Large Language Models
Programming over Thinking: Efficient and Robust Multi-Constraint Planning
ReCode: Reinforcing Code Generation with Reasoning-Process Rewards
Late Code Chunking: A Code Chunking Strategy for Repository-Level Code Completion
CLAOCS-TX: Cross-Lingual Triplet Extraction with Aspect-Opinion-Aware Code-Switched Prompting and LLM-Guided Contrastive Distillation
HSGraphAgent: Knowledge-Graph-Guided Large Language Models for Harmonized System Code Classification
Lost in the Mix: Evaluating LLM Understanding of Code-Switched Text
LogicEval: A Systematic Framework for Evaluating Automated Repair Techniques for Logical Vulnerabilities in Real-World Software
OLA: Output Language Alignment in Code-Switched LLM Interactions
GitChameleon 2.0: Evaluating AI Code Generation Against Python Library Version Incompatibilities
One Single Hub Text Breaks CLIP: Identifying Vulnerabilities in Cross-Modal Encoders via Hubness
TokDrift: When LLM Speaks in Subwords but Code Speaks in Grammar