Reasoning on a Budget: A Survey of Adaptive and Controllable Test-Time Compute in LLMs

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.

SKOLR: Structured Koopman Operator Linear RNN for Time-Series Forecasting

Koopman operator theory provides a framework for nonlinear dynamical system analysis and time-series forecasting by mapping dynamics to a space of real-valued measurement functions, enabling a linear operator representation. Despite the advantage of linearity, the operator is generally infinite-dimensional. Therefore, the objective is to learn measurement functions that yield a tractable finite-dimensional Koopman operator approximation. In this work, we establish a connection between Koopman operator approximation and linear Recurrent Neural Networks (RNNs), which have recently demonstrated remarkable success in sequence modeling. We show that by considering an extended state consisting of lagged observations, we can establish an equivalence between a structured Koopman operator and linear RNN updates. Building on this connection, we present SKOLR, which integrates a learnable spectral decomposition of the input signal with a multilayer perceptron (MLP) as the measurement functions and implements a structured Koopman operator via a highly parallel linear RNN stack. Numerical experiments on various forecasting benchmarks and dynamical systems show that this streamlined, Koopman-theory-based design delivers exceptional performance.

One Demo Is All It Takes: Planning Domain Derivation with LLMs from A Single Demonstration

Pre-trained Large Language Models (LLMs) have shown promise in solving planning problems but often struggle to ensure plan correctness, especially for long-horizon tasks. Meanwhile, traditional robotic task and motion planning (TAMP) frameworks address these challenges more reliably by combining high-level symbolic search with low-level motion planning. At the core of TAMP is the planning domain, an abstract world representation defined through symbolic predicates and actions. However, creating these domains typically involves substantial manual effort and domain expertise, limiting generalizability. We introduce Planning Domain Derivation with LLMs (PDDLLM), a novel approach that combines simulated physical interaction with LLM reasoning to improve planning performance. The method reduces reliance on humans by inferring planning domains from a single annotated task-execution demonstration. Unlike prior domain-inference methods that rely on partially predefined or language descriptions of planning domains, PDDLLM constructs domains entirely from scratch and automatically integrates them with low-level motion planning skills, enabling fully automated long-horizon planning. PDDLLM is evaluated on over 1,200 diverse tasks spanning nine environments and benchmarked against six LLM-based planning baselines, demonstrating superior long-horizon planning performance, lower token costs, and successful deployment on multiple physical robot platforms.

Half Search Space is All You Need

Neural Architecture Search (NAS) is a powerful tool for automating architecture design. One-Shot NAS techniques, such as DARTS, have gained substantial popularity due to their combination of search efficiency with simplicity of implementation. By design, One-Shot methods have high GPU memory requirements during the search. To mitigate this issue, we propose to prune the search space in an efficient automatic manner to reduce memory consumption and search time while preserving the search accuracy. Specifically, we utilise Zero-Shot NAS to efficiently remove low-performing architectures from the search space before applying One-Shot NAS to the pruned search space. Experimental results on the DARTS search space show that our approach reduces memory consumption by 81% compared to the baseline One-Shot setup while achieving the same level of accuracy.

Simplifying Graph Transformers

Transformers have attained outstanding performance across various modalities, employing scaled-dot-product (SDP) attention mechanisms. Researchers have attempted to migrate Transformers to graph learning, but most advanced Graph Transformers are designed with major architectural differences, either integrating message-passing or incorporating sophisticated attention mechanisms. These complexities prevent the easy adoption of Transformer training advances. We propose three simple modifications to the plain Transformer to render it applicable to graphs without introducing major architectural distortions. Specifically, we advocate for the use of (1) simplified $L_2$ attention to measure the magnitude closeness of tokens; (2) adaptive root-mean-square normalization to preserve token magnitude information; and (3) a relative positional encoding bias with a shared encoder. Significant performance gains across a variety of graph datasets justify the effectiveness of our proposed modifications. Furthermore, empirical evaluation on the expressiveness benchmark reveals noteworthy realized expressiveness in the graph isomorphism.

Variation Matters: from Mitigating to Embracing Zero-Shot NAS Ranking Function Variation

Neural Architecture Search (NAS) is a powerful automatic alternative to manual design of a neural network. In the zero-shot version, a fast ranking function is used to compare architectures without training them. The outputs of the ranking functions often vary significantly due to different sources of randomness, including the evaluated architecture's weights' initialization or the batch of data used for calculations. A common approach to addressing the variation is to average a ranking function output over several evaluations. We propose taking into account the variation in a different manner, by viewing the ranking function output as a random variable representing a proxy performance metric. During the search process, we strive to construct a stochastic ordering of the performance metrics to determine the best architecture. Our experiments show that the proposed stochastic ordering can effectively boost performance of a search on standard benchmark search spaces.

InnerThoughts: Disentangling Representations and Predictions in Large Language Models

Large language models (LLMs) contain substantial factual knowledge which is commonly elicited by multiple-choice question-answering prompts. Internally, such models process the prompt through multiple transformer layers, building varying representations of the problem within its hidden states. Ultimately, however, only the hidden state corresponding to the final layer and token position are used to predict the answer label. In this work, we propose instead to learn a small separate neural network predictor module on a collection of training questions, that take the hidden states from all the layers at the last temporal position as input and outputs predictions. In effect, such a framework disentangles the representational abilities of LLMs from their predictive abilities. On a collection of hard benchmarks, our method achieves considerable improvements in performance, sometimes comparable to supervised fine-tuning procedures, but at a fraction of the computational cost.

Secure Federated Graph-Filtering for Recommender Systems

Recommender systems often rely on graph-based filters, such as normalized item-item adjacency matrices and low-pass filters. While effective, the centralized computation of these components raises concerns about privacy, security, and the ethical use of user data. This work proposes two decentralized frameworks for securely computing these critical graph components without centralizing sensitive information. The first approach leverages lightweight Multi-Party Computation and distributed singular vector computations to privately compute key graph filters. The second extends this framework by incorporating low-rank approximations, enabling a trade-off between communication efficiency and predictive performance. Empirical evaluations on benchmark datasets demonstrate that the proposed methods achieve comparable accuracy to centralized state-of-the-art systems while ensuring data confidentiality and maintaining low communication costs. Our results highlight the potential for privacy-preserving decentralized architectures to bridge the gap between utility and user data protection in modern recommender systems.

Path-of-Thoughts: Extracting and Following Paths for Robust Relational Reasoning with Large Language Models

Large language models (LLMs) possess vast semantic knowledge but often struggle with complex reasoning tasks, particularly in relational reasoning problems such as kinship or spatial reasoning. In this paper, we present Path-of-Thoughts (PoT), a novel framework designed to tackle relation reasoning by decomposing the task into three key stages: graph extraction, path identification, and reasoning. Unlike previous approaches, PoT efficiently extracts a task-agnostic graph that identifies crucial entities, relations, and attributes within the problem context. Subsequently, PoT identifies relevant reasoning chains within the graph corresponding to the posed question, facilitating inference of potential answers. Experimental evaluations on four benchmark datasets, demanding long reasoning chains, demonstrate that PoT surpasses state-of-the-art baselines by a significant margin (maximum 21.3%) without necessitating fine-tuning or extensive LLM calls. Furthermore, as opposed to prior neuro-symbolic methods, PoT exhibits improved resilience against LLM errors by leveraging the compositional nature of graphs.

Hint Marginalization for Improved Reasoning in Large Language Models

Large Language Models (LLMs) have exhibited an impressive capability to perform reasoning tasks, especially if they are encouraged to generate a sequence of intermediate steps. Reasoning performance can be improved by suitably combining multiple LLM responses, generated either in parallel in a single query, or via sequential interactions with LLMs throughout the reasoning process. Existing strategies for combination, such as self-consistency and progressive-hint-prompting, make inefficient usage of the LLM responses. We present Hint Marginalization, a novel and principled algorithmic framework to enhance the reasoning capabilities of LLMs. Our approach can be viewed as an iterative sampling strategy for forming a Monte Carlo approximation of an underlying distribution of answers, with the goal of identifying the mode the most likely answer. Empirical evaluation on several benchmark datasets for arithmetic reasoning demonstrates the superiority of the proposed approach.