Datarus-R1: An Adaptive Multi-Step Reasoning LLM for Automated Data Analysis

• URL: http://arxiv.org/abs/2508.13382v1
• Date: Mon, 18 Aug 2025 21:58:18 GMT
• Title: Datarus-R1: An Adaptive Multi-Step Reasoning LLM for Automated Data Analysis
• Authors: Ayoub Ben Chaliah, Hela Dellagi,
• Abstract summary: We present Datarus-R1-14B, a language model fine-tuned from Qwen 2.5-14B-Instruct to act as a virtual data analyst and graduate-level problem solver.<n>Datarus is trained not on isolated question-answer pairs but on full analytical trajectories including reasoning steps, code execution, error traces, self-corrections, and final conclusions.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We present Datarus-R1-14B, a 14 B-parameter open-weights language model fine-tuned from Qwen 2.5-14B-Instruct to act as a virtual data analyst and graduate-level problem solver. Datarus is trained not on isolated question-answer pairs but on full analytical trajectories including reasoning steps, code execution, error traces, self-corrections, and final conclusions, all captured in a ReAct-style notebook format spanning finance, medicine, numerical analysis, and other quantitative domains. Our training pipeline combines (i) a trajectory-centric synthetic data generator that yielded 144 000 tagged notebook episodes, (ii) a dual-reward framework blending a lightweight tag-based structural signal with a Hierarchical Reward Model (HRM) that scores both single-step soundness and end-to-end coherence, and (iii) a memory-optimized implementation of Group Relative Policy Optimization (GRPO) featuring KV-cache reuse, sequential generation, and reference-model sharding. A cosine curriculum smoothly shifts emphasis from structural fidelity to semantic depth, reducing the format collapse and verbosity that often plague RL-aligned LLMs. A central design choice in Datarus is it dual reasoning interface. In agentic mode the model produces ReAct-tagged steps that invoke Python tools to execute real code; in reflection mode it outputs compact Chain-of-Thought (CoT) traces delimited by <think> and <answer> tags. On demanding postgraduate-level problems, Datarus exhibits an "AHA-moment" pattern: it sketches hypotheses, revises them once or twice, and converges avoiding the circular, token-inflating loops common to contemporary systems. Across standard public benchmarks Datarus surpasses similar size models and even reaches the level of larger reasoning models such as QwQ-32B achieving up to 30% higher accuracy on AIME 2024/2025 and LiveCodeBench while emitting 18-49% fewer tokens per solution.

Related papers

• Stacked from One: Multi-Scale Self-Injection for Context Window Extension [69.24689919827817]
  modelname is a novel framework based on multi-grained context compression and query-aware information acquisition.<n>modelnameachieves performance superior or comparable to strong baselines.
  arXiv  Detail & Related papers  (2026-03-05T03:16:16Z)
• ParEVO: Synthesizing Code for Irregular Data: High-Performance Parallelism through Agentic Evolution [13.109726609738749]
  ParEVO is a framework designed to synthesize high-performance parallel algorithms for irregular data.<n>On the ParEval benchmark, ParEVO achieves an average 106x speedup, and a robust 13.6x speedup on complex irregular graph problems.
  arXiv  Detail & Related papers  (2026-03-03T01:41:07Z)
• Test-Time Scaling with Diffusion Language Models via Reward-Guided Stitching [66.39914384073145]
  We propose a self-consistency framework that turns cheap diffusion-sampled reasoning into a reusable pool of step-level candidates.<n>We find that step-level recombination is most beneficial on harder problems.<n>Our training-free framework improves average accuracy by up to 2 across six math and coding tasks.
  arXiv  Detail & Related papers  (2026-02-26T11:08:39Z)
• Scaling LLM Speculative Decoding: Non-Autoregressive Forecasting in Large-Batch Scenarios [76.85739138203014]
  We present SpecFormer, a novel architecture that accelerates unidirectional and attention mechanisms.<n>We demonstrate that SpecFormer achieves lower training demands and reduced computational costs.
  arXiv  Detail & Related papers  (2025-11-25T14:20:08Z)
• Every Step Counts: Decoding Trajectories as Authorship Fingerprints of dLLMs [63.82840470917859]
  We show that the decoding mechanism of dLLMs can be used as a powerful tool for model attribution.<n>We propose a novel information extraction scheme called the Directed Decoding Map (DDM), which captures structural relationships between decoding steps and better reveals model-specific behaviors.
  arXiv  Detail & Related papers  (2025-10-02T06:25:10Z)
• Provably Efficient Online RLHF with One-Pass Reward Modeling [59.30310692855397]
  We propose a one-pass reward modeling method that does not require storing the historical data and can be computed in constant time.<n>We provide theoretical guarantees showing that our method improves both statistical and computational efficiency.<n>We conduct experiments using Llama-3-8B-Instruct and Qwen2.5-7B-Instruct models on the Ultrafeedback-binarized and Mixture2 datasets.
  arXiv  Detail & Related papers  (2025-02-11T02:36:01Z)
• Representation Similarity: A Better Guidance of DNN Layer Sharing for Edge Computing without Training [3.792729116385123]
  We propose a new model merging scheme by sharing representations at the edge, guided by representation similarity S. We show that S is extremely highly correlated with merged model's accuracy with Pearson Correlation Coefficient |r| > 0.94 than other metrics.
  arXiv  Detail & Related papers  (2024-10-15T03:35:54Z)
• NeRCC: Nested-Regression Coded Computing for Resilient Distributed Prediction Serving Systems [18.85527080950587]
  NeRCC is a general straggler-resistant framework for approximate coded computing. NeRCC accurately approximates the original predictions in a wide range of stragglers, outperforming the state-of-the-art by up to 23%.
  arXiv  Detail & Related papers  (2024-02-06T20:31:15Z)
• End-to-End Supervised Multilabel Contrastive Learning [38.26579519598804]
  Multilabel representation learning is recognized as a challenging problem that can be associated with either label dependencies between object categories or data-related issues. Recent advances address these challenges from model- and data-centric viewpoints. We propose a new end-to-end training framework -- dubbed KMCL -- to address the shortcomings of both model- and data-centric designs.
  arXiv  Detail & Related papers  (2023-07-08T12:46:57Z)
• Single-Stage Visual Relationship Learning using Conditional Queries [60.90880759475021]
  TraCQ is a new formulation for scene graph generation that avoids the multi-task learning problem and the entity pair distribution. We employ a DETR-based encoder-decoder conditional queries to significantly reduce the entity label space as well. Experimental results show that TraCQ not only outperforms existing single-stage scene graph generation methods, it also beats many state-of-the-art two-stage methods on the Visual Genome dataset.
  arXiv  Detail & Related papers  (2023-06-09T06:02:01Z)
• Disentanglement via Latent Quantization [60.37109712033694]
  In this work, we construct an inductive bias towards encoding to and decoding from an organized latent space. We demonstrate the broad applicability of this approach by adding it to both basic data-re (vanilla autoencoder) and latent-reconstructing (InfoGAN) generative models.
  arXiv  Detail & Related papers  (2023-05-28T06:30:29Z)
• Disentangled Recurrent Wasserstein Autoencoder [17.769077848342334]
  recurrent Wasserstein Autoencoder (R-WAE) is a new framework for generative modeling of sequential data. R-WAE disentangles the representation of an input sequence into static and dynamic factors. Our models outperform other baselines with the same settings in terms of disentanglement and unconditional video generation.
  arXiv  Detail & Related papers  (2021-01-19T07:43:25Z)
• AvgOut: A Simple Output-Probability Measure to Eliminate Dull Responses [97.50616524350123]
  We build dialogue models that are dynamically aware of what utterances or tokens are dull without any feature-engineering. The first model, MinAvgOut, directly maximizes the diversity score through the output distributions of each batch. The second model, Label Fine-Tuning (LFT), prepends to the source sequence a label continuously scaled by the diversity score to control the diversity level. The third model, RL, adopts Reinforcement Learning and treats the diversity score as a reward signal.
  arXiv  Detail & Related papers  (2020-01-15T18:32:06Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.