Inference-Time Adaptation of Language Models

Latest papers

24 papers3 papers in the last four weeks, against 2 the four weeks before. 0.0% of all new papers.

TIPS: A Tiny Inference-time Policy for Scaling in Diffusion Language Models
Oct 15, 2026Ya-Shi Zhang, Jian Tang, Pierre-André Noël +1Diffusion Language Model Decoding Inference-Time Adaptation of Language Models
ServiceNow AI Research · Mila - Quebec AI Institute · HEC Montréal
Self-Generated Feedback Destabilizes Test-Time Training: A Causal Decomposition of Long-Horizon Adaptation
Oct 4, 2026Cheng Luo, Bing Li, Bernard GhanemTest-Time Adaptation Continual Test-Time Adaptation
King Abdullah University of Science and Technology (KAUST)
Dynamic Routing as a New Dimension for Test-time Versatility of LLMs
Oct 3, 2026Michal Štefánik, Marek Kadlčík, Josef Kuchař +1Inference-Time Adaptation of Language Models LLM Routing
TransformersClub, Faculty of Informatics, Masaryk University · R&D Centre for Large Language Models, National Institute of Informatics, Japan
Test-time Calibration Learning for Large Language Model Reasoning
Oct 2, 2026Zizhuo Zhang, Xiong Peng, Jingwei Sun +4Inference-Time Adaptation of Language Models Language Model Calibration
TMLR Group, Department of Computer Science, Hong Kong Baptist University · Huawei Noah’s Ark Lab.
Beyond Confidence: Stability-Aware Test-Time Adaptation for LLM Reasoning
Sep 10, 2026Bincheng Gu, Min Gao, Zongwei Wang +3Test-Time Adaptation LLM Reliability
Key Laboratory of Dependable Service Computing in Cyber Physical Society, Chongqing University, China · King’s College London, London, UK · Griffith University, Brisbane, Australia
Safin-1: Safety from Within through Memory-Native State Evolution
Aug 31, 2026Ming Zhang, Kaisen Yang, Shu Yu +15Memory-Augmented Language Models LLM Safety Alignment
Shanghai AI Laboratory
Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching
Aug 10, 2026Kristian Schwethelm, Daniel Rueckert, Georgios KaissisLLM Inference LLM Inference Scheduling
Chair for AI in Healthcare and Medicine, Technical University of Munich (TUM), Germany · Department of Computing, Imperial College London, UK · Munich Center for Machine Learning (MCML), Germany +1
Strengthening Target-Language Features: SAE-Based Steering for Multilingual Inference
Aug 5, 2026Hongsheng Wang, Philipp KoehnMultilingual Language Models Language Model Steering
Johns Hopkins University, Baltimore / USA
SVR: Self-Verifying Refinement via Joint Verdict-Confidence Reinforcement Learning for Adaptive Test-Time Compute
Jul 30, 2026Hongyu Chen, Liang Lin, Guangrun WangLanguage Model Self-Verification RL for Language Model Reasoning
1Sun Yat-sen University · 2Guangdong Key Laboratory of Big Data Analysis and Processing · 3X-Era AI Lab
Frustratingly Simple Black-Box Adaptation of Language Models via Logit Bias
Jul 24, 2026Ofek I. Cohen, Lior Shani, Aviv Rosenberg +3Black-Box Optimization Domain Adaptation for LLMs
Tel Aviv University · Google Research · Columbia University
From Application-Layer Simulation to Native Meta-Architecture: Structural Tension as an Endogenous Driver for Heterogeneous AI Evolution
Jul 7, 2026Heting MaoLLM Alignment Memory-Augmented Language Models
Shanghai Lixin University of Accounting and Finance
Test-Time Training with Next-Token Prediction
Jun 19, 2026Xuan Ouyang, Zefan Cai, Junjie HuLong-Context Language Modeling Next-Token Prediction
University of Wisconsin–Madison
FAAST: Forward-Only Associative Learning via Closed-Form Fast Weights for Test-Time Supervised Adaptation
May 6, 2026Guangsheng Bao, Hongbo Zhang, Han Cui +4Efficient Neural Network Inference Test-Time Adaptation
Zhejiang University · Westlake University · Shanghai Polytechnic University +1
From Weights to Activations: Is Steering the Next Frontier of Adaptation?
Apr 15, 2026Simon Ostermann, Daniil Gurgurov, Tanja Baeumel +4Online Language Model Adaptation Language Model Steering
Saarland University · German Research Center for Artificial Intelligence (DFKI) · Centre for European Research in Trusted AI (CERTAIN) +6
Self-Calibrating Language Models via Test-Time Discriminative Distillation
Mar 18, 2026Mohamed Rissal Hedna, Jan Strich, Martin Semmann +1Inference-Time Adaptation of Language Models Language Model Calibration
Hub of Computing and Data Science (HCDS), University of Hamburg, Germany
Preconditioned Test-Time Adaptation for Out-of-Distribution Debiasing in Narrative Generation
Mar 14, 2026Hanwen Shen, Ting Ying, Jiajie Lu +1Narrative Generation Language Model Debiasing
Laboratory for Artificial Intelligence in Mathematics Education, Stevens Institute of Technology · Independent Researcher · NLP2CT Lab, Department of Computer and Information Science, University of Macau
Summarize Before You Speak with ARACH: A Training-Free Inference-Time Plug-In for Enhancing LLMs via Global Attention Reallocation
Mar 10, 2026Jingtao Wang, Yucong Wang, Jun Ding +2LLM Inference Inference-Time Adaptation of Language Models
Meakins-Christe Laboratories, Research Institute of McGill University Health Centre, 1001 Decarie Blvd, Montreal, H4A 3J1, Quebec, Canada · Department of Medicine, Division of Experimental Medicine, McGill University, 1001 Decarie Blvd, Montreal, H4A 3J1, Quebec, Canada · The College of Computer Science and Technology, Zhejiang Gongshang University, Hangzhou, China +4
Learning Adaptive LLM Decoding
Mar 10, 2026Chloe H. Su, Zhe Ye, Samuel Tenka +3Language Model Decoding Inference-Time Adaptation of Language Models
Department of Computer Science, Harvard University, Boston, USA · Kempner Institute, Harvard University, Cambridge, MA, USA · UC Berkeley, CA, USA +1
Test-Time Adaptation via Many-Shot Prompting: Benefits, Limits, and Pitfalls
Mar 6, 2026Shubhangi Upasani, Chen Wu, Jay Rainton +4Inference-Time Adaptation of Language Models LLM Prompting
SambaNova Systems, Inc · Stanford University
N-gram Injection into Transformers for Dynamic Language Model Adaptation in Handwritten Text Recognition
Mar 4, 2026Florent Meyer, Laurent Guichard, Yann Soullard +3Inference-Time Adaptation of Language Models Language Modeling
ANTAI, Rennes, France · Univ Rennes, CNRS, IRISA - UMR 6074, Rennes, France
Test-Time Meta-Adaptation with Self-Synthesis
Mar 3, 2026Zeyneb N. Kaya, Nick RuiInference-Time Adaptation of Language Models Test-Time Adaptation
Stanford University
PonderLM-3: Adaptive Token-Wise Pondering with Differentiable Masking
Mar 2, 2026He Li, Feichen Song, Boyi Zeng +4Adaptive Computation Inference-Time Adaptation of Language Models
LUMIA Lab, School of Artificial Intelligence, Shanghai Jiao Tong University · Shanghai AI Laboratory · Shanghai Jiao Tong University +1
AdaPonderLM: Gated Pondering Language Models with Token-Wise Adaptive Depth
Mar 2, 2026Shixiang Song, He Li, Zitong Wang +6Recurrent Language Models Inference-Time Adaptation of Language Models
LUMIA Lab, School of Artificial Intelligence, Shanghai Jiao Tong University · Shanghai Innovation Institute · Sun Yat-sen University +2
Words & Weights: Streamlining Multi-Turn Interactions via Co-Adaptation
Mar 2, 2026Chenxing Wei, Hong Wang, Ying He +4Inference-Time Adaptation of Language Models Inference-Time Language Model Alignment
Shenzhen University · Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ) · Bytedance +4

Oct 15, 2026J/K move · Enter open · S save

TIPS: A Tiny Inference-time Policy for Scaling in Diffusion Language Models

Ya-Shi Zhang, Jian Tang, Pierre-André Noël, Oleksiy Ostapenko

ServiceNow AI Research · Mila - Quebec AI Institute · HEC Montréal

Diffusion Language Model Decoding Inference-Time Adaptation of Language Models Inference-Time Scaling Masked Diffusion Models

Open paper AskPDF arXiv

Masked diffusion language models expose an inference-time choice: which positions become visible after each denoising pass. Many existing decoding methods use hand-designed rules or require visibility to grow monotonically. We present a

\textbf{T}

iny

\textbf{I}

nference-time

\textbf{P}

olicy for

\textbf{S}

caling (

\textbf{TIPS}

), an approximately

600

K-parameter controller that learns both unmasking and remasking while keeping the underlying language model frozen. TIPS uses visibility-aware confidence features and a shallow transformer to select which tokens should be visible, while a fixed schedule controls the expected visible fraction. We train the controller through supervised warmup followed by reinforcement learning from terminal correctness rewards. On GSM8K and MATH-500 with LLaDA-8B-Instruct, the full-canvas and two-block TIPS variants achieve

58.46\%

and

57.97\%

micro-averaged accuracy at a budget of

32

decoder evaluations, exceeding the strongest baseline in our main comparison by

12.1

and

11.6

percentage points, respectively. Both variants are trained at a budget of

32

evaluations and remain competitive at higher budgets without retraining. These results demonstrate that learning when to reveal and reconsider tokens can improve the accuracy-compute trade-off without updating the decoder backbone.