Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,657 papers · 148 categories

Trend · papers per month

19385675 · Jun 202019922001200920172026
48 results for multi-turn reasoning

Enhances math problem-solving models with multi-turn preference learning.

problem Improving mathematical problem-solving capabilities of large language models.
method Introduces a multi-turn direct preference learning framework for tool-integrated mathematical reasoning tasks.
result Significant performance improvements in model accuracy on math datasets.

This study compares hierarchical and non-hierarchical models for open-domain multi-turn dialog generation.

problem Which kind of models (hierarchical or non-hierarchical) is better for open-domain multi-turn dialog generation?
method Systematically compared nearly all representative hierarchical and non-hierarchical models over the same experimental settings.
result Nearly all hierarchical models are worse than non-hierarchical models in open-domain multi-turn dialog generation, except for HRAN.

ReOPD uses pre-collected teacher trajectories to distill knowledge from multi-turn interactions.

problem The cost of fully online on-policy distillation for multi-turn interactions.
method ReOPD, an off-environment alternative that reuses pre-collected teacher trajectories as replayed prefixes, addressing the prefix trap and distribution shift.
result ReOPD preserves or improves OPD-level accuracy, uses zero tool calls, and is at least 4imes imes faster per training step.

Foundation models struggle with multi-turn exploration but can learn through regular summaries.

problem Foundation models struggle with multi-turn exploration in dynamic environments.
method Implemented a text-based version of the Alchemy environment to test multi-trial learning. Prompting models to summarize their observations at regular intervals enabled them to improve across trials and adapt to changes.
result Foundation models can improve through regular summaries, enabling multi-trial learning and adaptation.

We present Meena, a multi-turn open-domain chatbot trained end-to-end on data mined and filtered from public domain social media conversations. This 2.6B parameter neural network is simply trained to minimize perplexity of the next token. We also propose a human evaluation metric called Sensibleness and Specificity Ave…

2020-01-27abs ↗pdf ↗

Neural dialogue models, despite their successes, still suffer from lack of relevance, diversity, and in many cases coherence in their generated responses. These issues can attributed to reasons including (1) short-range model architectures that capture limited temporal dependencies, (2) limitations of the maximum likel…

2019-07-26abs ↗pdf ↗

Efficient algorithm for learning from indirect feedback in complex decision-making scenarios.

problem Learning from indirect feedback in realistic scenarios with personalized mechanisms.
method IGW algorithm for policy optimization, extending reward-estimator construction from single-step to multi-step.
result Achieves sublinear regret guarantee for contextual episodic MDPs with personalized feedback.

HabitatAgent offers a multi-agent system for transparent housing consultation.

problem Opaque reasoning and brittle multi-constraint handling in housing recommendation systems.
method HabitatAgent is a multi-agent architecture with specialized roles for memory, retrieval, generation, and validation.
result HabitatAgent achieves 95% accuracy in real user consultation scenarios, significantly outperforming a strong baseline.

Most existing text-to-image synthesis tasks are static single-turn generation, based on pre-defined textual descriptions of images. To explore more practical and interactive real-life applications, we introduce a new task - Interactive Image Editing, where users can guide an agent to edit images via multi-turn textual …

2018-12-20abs ↗pdf ↗

StableLM 2 1.6B is a new language model series with detailed evaluations and performance metrics.

problem Improving language models with smaller sizes and better performance.
method Detailed data and training procedure for StableLM 2 1.6B, including zero- and few-shot benchmarks and multilingual evaluations.
result StableLM 2 1.6B is the state-of-the-art open model under 2B parameters.

BED-LLM uses Bayesian experimental design to improve LLMs' information gathering.

problem Improving LLMs' ability to gather information adaptively.
method Iteratively choosing questions to maximize expected information gain using a probabilistic model.
result BED-LLM achieves substantial performance gains compared to other adaptive design strategies.

LaTRO optimizes latent reasoning in LLMs without external reward.

problem Training LLMs to perform complex reasoning tasks.
method Formulates reasoning as latent distribution sampling and optimizes via variational approaches.
result LLMs improve reasoning and evaluation quality through self-improvement.

Auto-CEI improves LLM reasoning by balancing assertiveness and conservativeness.

problem Hallucinations and laziness in LLM reasoning tasks.
method Expert Iteration explores reasoning trajectories, guiding incorrect paths back on track and promoting appropriate 'I don't know' responses.
result Auto-CEI achieves superior alignment in logical reasoning, mathematics, and planning tasks.

A new method for math reasoning that allows for iterative correction.

problem Standard reasoning models commit to each token and cannot recover from early errors.
method Generative framework with latent thought vectors for iterative self-correction.
result 30 rethinking iterations surpass baselines with 15 times more parameters.

Inferring new facts from existing knowledge graphs (KG) with explainable reasoning processes is a significant problem and has received much attention recently. However, few studies have focused on relation types unseen in the original KG, given only one or a few instances for training. To bridge this gap, we propose Co…

2019-06-13abs ↗pdf ↗

Transformers learn multi-step reasoning through gradient descent.

problem Understanding how transformers solve symbolic multi-step reasoning tasks.
method Theoretical analysis of gradient descent dynamics and multi-phase training.
result Trained one-layer transformers can solve both backward and forward reasoning tasks with generalization guarantees.

Early stopping methods reduce unnecessary reasoning steps in LLMs by monitoring uncertainty signals.

problem LLMs sometimes generate unnecessary reasoning steps, especially under uncertainty.
method Statistically principled early stopping methods that monitor uncertainty signals during generation.
result Uncertainty-aware early stopping improves efficiency and reliability in LLM reasoning, especially in math reasoning.

Achieving artificial visual reasoning - the ability to answer image-related questions which require a multi-step, high-level process - is an important step towards artificial general intelligence. This multi-modal task requires learning a question-dependent, structured reasoning process over images from language. Stand…

2017-07-10abs ↗pdf ↗

FinTradeBench benchmarks LLMs for financial reasoning combining company fundamentals and market signals.

problem Challenges in evaluating financial reasoning models for LLMs.
method Developed a benchmark integrating company fundamentals and trading signals, using a calibration-then-scaling framework.
result Clear performance gap between LLMs, retrieval improves reasoning over textual fundamentals but not trading signals.

Hybrid framework injects TSLM insights into GRLM for robust time-series reasoning.

problem Lack of domain-specific knowledge in large language models for time-series reasoning.
method Hybrid knowledge-injection framework combining RLVR for efficient knowledge transfer.
result Consistently outperforms existing models by 7.9%-26.1% on multivariate time-series benchmarks.

Neural Logic Reasoning integrates deep learning and symbolic logic for better prediction tasks.

problem Lack of cognitive reasoning in deep neural networks limits their ability to solve complex prediction tasks.
method Proposes Logic-Integrated Neural Network (LINN) that learns logical operations and conducts propositional logical reasoning.
result LINN significantly outperforms state-of-the-art recommendation models in Top-K recommendation.

Forward-prediction models enhance physical reasoning, but only for specific tasks.

problem Improving physical reasoning in complex tasks involving many objects.
method Incorporated forward-prediction models into simple physical-reasoning agents and evaluated their performance on the PHYRE benchmark.
result Forward-prediction models improve physical-reasoning performance, especially on complex tasks, but generalization to new task templates is challenging.

RACER optimizes LLM-as-judge accuracy with dynamic reasoning selection.

problem Balancing reasoning accuracy with computational cost in LLM-as-judge settings.
method Formulates routing as a constrained distributionally robust optimization problem, accounting for distribution shift via KL-divergence uncertainty set.
result RACER achieves superior accuracy-cost trade-offs under distribution shift.

VTA combines verbal and latent reasoning for accurate stock time-series forecasts.

problem Challenges in combining textual analysis with time-series data for financial forecasting.
method Converts stock price data into textual annotations, optimizes reasoning trace using inverse MSE, conditions time-series model outputs on reasoning attributes.
result VTA achieves state-of-the-art forecasting accuracy and interpretable reasoning traces.

Fractured Sampling improves LLM reasoning efficiency by truncating CoT trajectories.

problem Efficiently scaling reasoning in large language models with limited tokens.
method Integrating truncated Chain-of-Thought (CoT) with Fractured Sampling across multiple dimensions.
result Fractured Sampling achieves superior accuracy-cost trade-offs compared to full CoT.

Few-shot visual reasoning model learns analogical relationships from small data.

problem Training deep models on few samples for visual reasoning tasks.
method Meta-analogical contrastive learning to enforce structural similarity between training and test samples.
result Method outperforms state-of-the-art on RAVEN dataset with scarce training data.

Study reveals how depth of reasoning affects generalization in models.

problem Understanding scaling behavior of generalization with CoT depth.
method Theoretical model of CoT in linear regression using random matrix theory.
result Sharp phase transition between exponential and polynomial improvement, saturation, and overthinking.

Enhances large language models' reasoning through simpler off-policy reinforcement learning.

problem Improving large language models' ability to reason and solve problems.
method EM Policy Gradient, optimizing expected return over reasoning trajectories using Expectation-Maximization (EM) optimization.
result Achieves comparable or slightly superior performance to state-of-the-art methods on reasoning datasets, with additional cognitive behaviors.

Self-supervised skip-tree training improves mathematical reasoning in language models.

problem Improving logical reasoning in language models for formal mathematics.
method Self-supervised language modeling on mathematical formulas, skip-tree task.
result Models trained on skip-tree task outperform standard models in mathematical reasoning tasks.

Theoretical work shows integrating coherent reasoning improves LLM performance and error correction.

problem Improving reasoning and error correction in large language models (LLMs) with few-shot prompting.
method Theoretical analysis and sensitivity experiments on transformer behavior with coherent reasoning and corrupted demonstrations.
result The transformer gains better error correction ability and more accurate predictions when coherent reasoning is integrated.

R2-B2 optimizes game interactions with recursive reasoning.

problem Optimizing interactions between boundedly rational agents with unknown payoff functions.
method Recursive Reasoning-Based Bayesian Optimization (R2-B2) for repeated games.
result R2-B2 achieves faster asymptotic convergence to no regret than non-recursive methods.

This work investigates how multi-round reasoning improves LLM performance.

problem Improving problem-solving abilities in complex tasks with LLMs.
method Investigates approximation, learnability, and generalization properties of multi-round auto-regressive models.
result Transformers with finite context windows are universal approximators for Turing-computable functions and can approximate any Turing-computable sequence-to-sequence function through multi-round reasoning.

MXGNet tackles visual reasoning tasks using graph neural networks.

problem Abstract reasoning, especially in the visual domain, is challenging for AI.
method Combines object-level representations, graph neural networks, and multiplex graphs.
result Achieves state-of-the-art accuracy on Euler Diagram Syllogisms and outperforms state-of-the-art models on RPM datasets.

This study explains RL training dynamics in LLMs, focusing on token-level optimization and reasoning pattern reshaping.

problem Understanding the training dynamics of RL in LLMs to improve their reasoning capabilities.
method Empirical analysis and theoretical modeling of RL training process, focusing on reasoning patterns and token optimization.
result RL primarily optimizes a sparse subset of critical tokens, reshaping reasoning pattern distributions and affecting model performance.

The paper studies how search and distillation improve reasoning in large language models.

problem Improving reasoning capabilities of large language models.
method Viewing chain-of-thought generation as a metastable Markov process, proving benefits of search and distillation.
result Search protocol rewards sparse edges, reducing the expected number of steps to reach different clusters.