REVIEW 10 cited by
Reasoning on a Budget: A Survey of Adaptive and Controllable Test-Time Compute in LLMs
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Reasoning on a Budget: A Survey of Adaptive and Controllable Test-Time Compute in LLMs
read the original abstract
Large language models (LLMs) have rapidly progressed into general-purpose agents capable of solving a broad spectrum of tasks. However, current models remain inefficient at reasoning: they apply fixed inference-time compute regardless of task complexity, often overthinking simple problems while underthinking hard ones. This survey presents a comprehensive review of efficient test-time compute (TTC) strategies, which aim to improve the computational efficiency of LLM reasoning. We introduce a two-tiered taxonomy that distinguishes between L1-controllability, methods that operate under fixed compute budgets, and L2-adaptiveness, methods that dynamically scale inference based on input difficulty or model confidence. We benchmark leading proprietary LLMs across diverse datasets, highlighting critical trade-offs between reasoning performance and token usage. Compared to prior surveys on efficient reasoning, our review emphasizes the practical control, adaptability, and scalability of TTC methods. Finally, we discuss emerging trends such as hybrid thinking models and identify key challenges for future work towards making LLMs more computationally efficient, robust, and responsive to user constraints.
Forward citations
Cited by 10 Pith papers
-
DART: Draft-Agreement Routing for Training-Free Adaptive Thinking Budgets in Hybrid Reasoning Models
DART is a training-free router that accepts direct answers on draft agreement and allocates thinking budgets via draft entropy on disagreement, reporting accuracy gains and token reductions on math and code benchmarks...
-
Not All Errors Are Equal: Consequence-Aware Reasoning Compute Allocation
Consequence-aware scheduler using an issue-text predictor routes more compute to high-cost failures and cuts cost-weighted loss by 22-33% versus difficulty-based allocation on SWE-bench tasks.
-
SVR: Self-Verifying Refinement via Joint Verdict-Confidence Reinforcement Learning for Adaptive Test-Time Compute
Learned verdict–confidence self-verification can control multi-turn refinement and adaptive test-time compute without external correctness feedback at inference.
-
ARES: Adaptive Reasoning-Effort Steering for PPA- and Cost-Aware RTL Optimization with LLM Agents
Escalating an LLM agent's reasoning effort only when cheap edits stall improves RTL power-area-delay by 23–27% versus 16–23% for the best fixed effort, at equal dollar cost.
-
Budget-Aware LLM Discovery via Cost-Calibrated Frontier Utility
Cost-blind LLM discovery controllers can waste almost all attainable quality under heterogeneous action costs; CostAda’s cost-calibrated frontier utility recovers budget efficiency and final quality across eight benchmarks.
-
Budget-Aware LLM Discovery via Cost-Calibrated Frontier Utility
Cost-calibrated frontier utility — progress divided by realized token cost and weighted by remaining budget — lets CostAda reach full-budget quality at half the budget in 12 of 16 benchmark–backbone pairs and achieve ...
-
Implicit Compression Regularization: Concise Reasoning via Internal Shorter Distributions in RL Post-Training
ICR creates a virtual shorter distribution from shortest correct on-policy responses to regularize RL post-training toward concise yet accurate reasoning, improving the accuracy-length Pareto frontier on math and know...
-
Stream-T1: Test-Time Scaling for Streaming Video Generation
Stream-T1 is a test-time scaling framework for streaming video generation using scaled noise propagation from history, reward pruning across short and long windows, and feedback-guided memory sinking to improve tempor...
-
Efficient Reasoning on the Edge
LoRA adapters, budget-forced GRPO, dynamic switching, parallel verification and FPTQuant enable practical chain-of-thought reasoning on quantized Qwen2.5-7B for edge devices.
-
Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Models
The paper unifies perspectives on Long CoT in reasoning LLMs by introducing a taxonomy, detailing characteristics of deep reasoning and reflection, and discussing emergence phenomena and future directions.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.