The reviewed record of science sign in
Pith

arxiv: 2408.05517 · v4 · pith:YQS2S3KO · submitted 2024-08-10 · cs.CL

SWIFT:A Scalable lightWeight Infrastructure for Fine-Tuning

Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:YQS2S3KOrecord.jsonopen to challenge →

classification cs.CL
keywords modelslargeswiftfine-tuningtrainingmllmsframeworkllms
0
0 comments X
read the original abstract

Recent development in Large Language Models (LLMs) and Multi-modal Large Language Models (MLLMs) have leverage Attention-based Transformer architectures and achieved superior performance and generalization capabilities. They have since covered extensive areas of traditional learning tasks. For instance, text-based tasks such as text-classification and sequence-labeling, as well as multi-modal tasks like Visual Question Answering (VQA) and Optical Character Recognition (OCR), which were previously addressed using different models, can now be tackled based on one foundation model. Consequently, the training and lightweight fine-tuning of LLMs and MLLMs, especially those based on Transformer architecture, has become particularly important. In recognition of these overwhelming needs, we develop SWIFT, a customizable one-stop infrastructure for large models. With support of over $300+$ LLMs and $50+$ MLLMs, SWIFT stands as the open-source framework that provide the most comprehensive support for fine-tuning large models. In particular, it is the first training framework that provides systematic support for MLLMs. In addition to the core functionalities of fine-tuning, SWIFT also integrates post-training processes such as inference, evaluation, and model quantization, to facilitate fast adoptions of large models in various application scenarios. With a systematic integration of various training techniques, SWIFT offers helpful utilities such as benchmark comparisons among different training techniques for large models. For fine-tuning models specialized in agent framework, we show that notable improvements on the ToolBench leader-board can be achieved by training with customized dataset on SWIFT, with an increase of 5.2%-21.8% in the Act.EM metric over various baseline models, a reduction in hallucination by 1.6%-14.1%, and an average performance improvement of 8%-17%.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 31 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. S1-VL: Scientific Multimodal Reasoning Model with Thinking-with-Images

    cs.CV 2026-04 unverdicted novelty 8.0

    S1-VL combines structured scientific reasoning with iterative image manipulation via code execution to reach state-of-the-art results on visual and scientific reasoning benchmarks.

  2. PInVerify: An Offline Embodied Benchmark for Active Instance Verification

    cs.CV 2026-05 unverdicted novelty 7.0

    PInVerify is a new offline embodied benchmark for active instance verification that supplies multi-view captures and 6-sector navigation topology, with MLLM baselines reaching 85.6% after fine-tuning but showing no re...

  3. K-FinHallu: A Hallucination Detection Benchmark for Multi-Turn RAG in Korean Finance

    cs.LG 2026-05 unverdicted novelty 7.0

    K-FinHallu is the first multi-turn Korean financial RAG hallucination benchmark; frontier LLMs struggle especially on justified abstention while an 8B fine-tuned model reaches competitive performance.

  4. Omni-Persona: Systematic Benchmarking and Improving Omnimodal Personalization

    cs.CV 2026-05 unverdicted novelty 7.0

    Omni-Persona benchmark with 18 tasks shows open-source models have audio-visual grounding gaps, RLVR narrows them but leads to conservative outputs, and scale or recall alone fail as diagnostics.

  5. HyperEyes: Dual-Grained Efficiency-Aware Reinforcement Learning for Parallel Multimodal Search Agents

    cs.LG 2026-05 unverdicted novelty 7.0

    HyperEyes uses a dual-grained RL framework with parallel tool actions and efficiency rewards to achieve 9.9% higher accuracy and 5.3x fewer tool calls than prior open-source multimodal agents.

  6. Rethinking Reasoning-Intensive Retrieval: Evaluating and Advancing Retrievers in Agentic Search Systems

    cs.CL 2026-05 unverdicted novelty 7.0

    BRIGHT-Pro and RTriever-Synth advance reasoning-intensive retrieval by adding multi-aspect evidence evaluation and aspect-decomposed synthetic training, with the fine-tuned RTriever-4B showing gains over its base model.

  7. Steadily moving semi-infinite fracture in plane poroelasticity

    physics.geo-ph 2026-04 unverdicted novelty 7.0

    A new coupled boundary integral method models steadily moving semi-infinite fractures in plane poroelasticity, solving for mechanical deformation and fluid exchange with verification on analytical test cases.

  8. KeyframeFace: Language-Driven Facial Animation via Semantic Keyframes

    cs.CV 2025-12 unverdicted novelty 7.0

    KeyframeFace uses LLM priors and semantic keyframe supervision in ARKit space to produce language-driven facial animations with improved fidelity and interpretability over continuous regression methods.

  9. PIAST: Rapid Prompting with In-context Augmentation for Scarce Training data

    cs.CL 2025-12 conditional novelty 7.0

    PIAST iteratively optimizes few-shot examples in prompts via Monte Carlo Shapley value estimation, outperforming prior automatic prompting methods and setting new SOTA on classification, simplification, and GSM8K with...

  10. Do Multimodal Large Language Models Need Reasoning to Classify Dementia from Speech?

    eess.AS 2026-06 unverdicted novelty 6.0

    DeTAiL adaptor framework extracts internal representations from reasoning MLLMs via nonlinear adaptor and RL to outperform baselines and text-rationale methods for speech-based dementia classification.

  11. Fine-grained Fragment Retrieval in Multi-modal Long-form Dialogues

    cs.CL 2026-06 unverdicted novelty 6.0

    Introduces FFR task, F2RVLM and FFRS models, and MLDR dataset for retrieving coherent multi-modal dialogue fragments, reporting superior performance on single-dialogue and corpus benchmarks.

  12. When Should Models Change Their Minds? Contextual Belief Management in Large Language Models

    cs.AI 2026-05 unverdicted novelty 6.0

    Introduces BeliefTrack benchmark diagnosing three CBM failures in LLMs and shows RL with belief-state rewards cuts failure rates by 70.9% while representation steering cuts them by 46.1%.

  13. Qwen-Image-Bench: From Generation to Creation in Text-to-Image Evaluation

    cs.CV 2026-05 unverdicted novelty 6.0

    Qwen-Image-Bench introduces a hierarchical creator-centric benchmark with 1000 prompts, 23 sub-capabilities, and a Q-Judger model that scores images on 56 verifiable facets to distinguish T2I models on fidelity and cr...

  14. Learning When to Think While Listening in Large Audio-Language Models

    cs.CL 2026-05 unverdicted novelty 6.0

    A wait-think-answer controller for LALMs is trained via SFT followed by six-reward DAPO, raising row-weighted accuracy from 67.6% to 70.3% and cutting post-endpoint thinking length by 14% on synthetic spoken QA while ...

  15. From Patches to Trajectories: Privileged Process Supervision for Software-Engineering Agents

    cs.SE 2026-05 unverdicted novelty 6.0

    P2T distills reference patches into a latent process graph and uses it to select shortest effective trajectory segments from teacher rollouts, yielding up to 10.8 point Pass@1 gains on SWE-bench Verified with 15% lowe...

  16. TimeSRL: Generalizable Time-Series Behavioral Modeling via Semantic RL-Tuned LLMs -- A Case Study in Mental Health

    cs.LG 2026-05 unverdicted novelty 6.0

    TimeSRL uses semantic abstractions from time-series data optimized via reinforcement learning to achieve better cross-dataset generalization than standard ML or LLM baselines in mental health prediction.

  17. Video Understanding Reward Modeling: A Robust Benchmark and Performant Reward Models

    cs.CV 2026-05 unverdicted novelty 6.0

    Introduces VURB benchmark and VUP-35K dataset to train discriminative and generative video reward models that achieve SOTA performance on VURB and VideoRewardBench.

  18. HyperEyes: Dual-Grained Efficiency-Aware Reinforcement Learning for Parallel Multimodal Search Agents

    cs.LG 2026-05 unverdicted novelty 6.0

    HyperEyes presents a parallel multimodal search agent using dual-grained efficiency-aware RL with a new TRACE reward and IMEB benchmark, claiming 9.9% higher accuracy and 5.3x fewer tool calls than prior open-source agents.

  19. JU\'A -- A Benchmark for Information Retrieval in Brazilian Legal Text Collections

    cs.IR 2026-04 accept novelty 6.0

    JU'A is a new heterogeneous benchmark for Brazilian legal IR that distinguishes retrieval methods and shows domain-adapted models excel on aligned subsets while BM25 stays competitive elsewhere.

  20. Logics-Parsing-Omni Technical Report

    cs.AI 2026-03 unverdicted novelty 6.0

    Omni Parsing framework converts complex multimodal signals into locatable, enumerable, and traceable structured knowledge via hierarchical detection, recognition, and interpreting with strict evidence alignment.

  21. Do Fine-Tuned LLMs Understand Vulnerabilities? An Investigation into the Semantic Trap

    cs.CR 2026-01 unverdicted novelty 6.0

    Fine-tuned decoder-only LLMs fall into a Semantic Trap on vulnerability detection, achieving high scores on unpaired normal code but failing on paired vulnerable-patched code, semantic perturbations, and gap analysis,...

  22. LLM-AutoDP: Automatic Data Processing via LLM Agents for Model Fine-tuning

    cs.LG 2026-01 unverdicted novelty 6.0

    LLM agents iteratively generate and optimize data processing strategies for fine-tuning, delivering over 80% win rates versus unprocessed data and 65% versus LLM-based AutoML baselines while cutting search time by up to 10x.

  23. Do Multimodal Large Language Models Need Reasoning to Classify Dementia from Speech?

    eess.AS 2026-06 unverdicted novelty 5.0

    DeTAiL uses internal representations from reasoning MLLMs via an adaptor and RL to outperform text-rationale methods and baselines for speech-based dementia classification on two datasets.

  24. AlloSpatial: Agentic Harness Framework for Spatial Reasoning in Foundation Models

    cs.AI 2026-06 unverdicted novelty 5.0

    AlloSpatial adds structured allocentric priors and a harness for tool-use and arbitration to improve spatial reasoning in foundation models, with 5-18% gains on VSI-Bench and MindCube in training-free settings and fur...

  25. BeLink: Biomedical Entity Linking Meets Generative Re-Ranking

    cs.CL 2026-05 unverdicted novelty 5.0

    BeLink applies set-wise instruction-tuning to generative LLMs at the re-ranking stage of biomedical entity linking, reporting 3-24% accuracy gains and reduced inference time versus prior methods.

  26. Steadily moving semi-infinite fracture in plane poroelasticity

    physics.geo-ph 2026-04 conditional novelty 5.0

    XEmbodied achieves SOTA on 18 embodied VQA benchmarks by fusing 3D geometric tokens and distilled physical cues into a 30B VLM with progressive curriculum training.

  27. Teach-and-Repeat: Accurately Extracting Operational Knowledge from Mobile Screen Demonstrations to Empower GUI Agents

    cs.AI 2026-06 unverdicted novelty 4.0

    Teach VLM extracts operational knowledge from screen trajectories and the Teach-and-Repeat paradigm uses it to improve downstream GUI agent task success rates.

  28. Internalize the Temperature: On-Policy Self-Distillation as Policy Reheater for Reinforcement Learning

    cs.CL 2026-05 unverdicted novelty 4.0

    TS-OPSD internalizes temperature via on-policy self-distillation to reheat entropy-collapsed RL policies in LLMs, providing stronger initialization for further training than continued RL or rollout temperature adjustment.

  29. Domain-Adaptive Dense Retrieval for Brazilian Legal Search

    cs.IR 2026-05 unverdicted novelty 4.0

    Mixed training of Qwen3-Embedding-4B on legal data plus SQuAD-pt yields higher average NDCG@10 (0.447), MRR@10 (0.595), and MAP@10 (0.308) across six Portuguese retrieval datasets than legal-only or base models, with ...

  30. XEmbodied: A Foundation Model with Enhanced Geometric and Physical Cues for Large-Scale Embodied Environments

    cs.CV 2026-04 unverdicted novelty 4.0

    XEmbodied is a foundation model that integrates 3D geometric and physical signals into VLMs using a 3D Adapter and Efficient Image-Embodied Adapter, plus progressive curriculum and RL post-training, to improve spatial...

  31. EasyVideoR1: Easier RL for Video Understanding

    cs.CV 2026-04 unverdicted novelty 4.0

    EasyVideoR1 delivers an optimized RL pipeline for video understanding in large vision-language models, achieving 1.47x throughput gains and aligned results on 22 benchmarks.