ReciteHelper v5 "Lotus"

ReciteHelper v5 “Lotus”

New features

• Introduces the FSRS-6 memory model, replacing the legacy SM-2-based review algorithm with recall-probability-driven scheduling
• Automatically prioritizes questions that are most likely to be forgotten and suggests the next review time
• Supports per-project scheduler personalization after sufficient review history has been collected
• Expands practice support to five question types: single-choice, fill-in-the-blank, true/false, term-definition, and essay questions
• Adds complete exam-set importing from PDF, TXT, HTML, and MHTML files
• Adds chapter-weighted exam generation, timed exams, automatic grading, answer review, and mistake importing
• Introduces a project-local vector knowledge base for semantic retrieval
• Adds a question assistant that retrieves relevant knowledge points and generates grounded AI explanations
• Adds three model-service options:
  • Direct DeepSeek + Qwen access
  • OpenRouter access with a single API key
  • ReciteHelper hosted service using an activation code
• Adds a model-service selection window for first-time setup
• Adds Resource Center support for discovering and importing shared learning resources
• Adds support for importing XuetangX webpages and exam content
• Adds configurable import and export destination paths

Enhancements

• Rebuilds the smart-review workflow around difficulty, stability, and retrievability instead of static easiness-factor ranking
• Preserves historical review data and automatically migrates legacy review records to the new scheduler
• Improves project creation with structured stages, clearer progress reporting, and better error feedback
• Refines chapter extraction and structured question generation prompts
• Improves fill-in-the-blank recognition and answer normalization
• Improves question generation quality across different source formats
• Improves knowledge-point matching with highlighted relevant text
• Keeps each project’s knowledge base, questions, and learning history together in local storage
• Automatically rebuilds knowledge-base vectors when switching to an incompatible embedding model
• Updates the quiz interface to display predicted recall, memory stability, and suggested review timing
• Simplifies first-time configuration by allowing API keys or activation codes to be entered directly in the application
• Adds configurable OpenRouter chat and embedding models
• Improves price estimation and peak-hour calculation
• Refines project import, export, loading, and compatibility handling
• Reorganizes the README with Chinese and English versions, clearer onboarding, real interface previews, and updated project links

Bug Fixes

• Fixed a startup failure when no model-service keys were configured by restoring a dedicated configuration window
• Fixed failures caused by a missing SBERT vocabulary file by adding a safe fallback
• Fixed legacy review records not being restored correctly during project deserialization
• Fixed time-blind review selection that repeatedly prioritized previously failed questions regardless of elapsed time
• Fixed inconsistent handling of true/false, fill-in-the-blank, and structured exam questions
• Fixed several project-path and import/export destination issues
• Fixed problems when identifying multiple exam papers from a single source file
• Fixed malformed or incomplete chapter structures produced during AI generation
• Fixed silent knowledge-base degradation when changing embedding providers
• Fixed configuration checks that previously required both DeepSeek and Qwen even when another model-service option was available
• Fixed several known crashes and compatibility issues in project creation and exam importing

Build / Test updates

• Refactors the application into separate Core, Application, Infrastructure, SharedKernel, and WPF projects
• Updates the primary solution to src/ReciteHelper.slnx
• Updates the build workflow to restore Git submodules recursively
• Migrates the application to C#, .NET 10, and the current WPF project structure
• Adds dependency-injection registrations for the FSRS scheduler and personalization service
• Removes the obsolete SM-2 services, XGBoost quality predictor, ONNX model, and training scripts
• Adds deterministic FSRS state updates, bounded scheduler parameters, legacy-history migration, and on-device personalization
• Verifies that the complete Release solution builds successfully after removing the legacy algorithm

Changelog

Full Changelog: v4...v5

Lotus rises from still water; every review brings knowledge back into bloom. Welcome to ReciteHelper 5.0!

ReciteHelper v4 "Calamus"

ReciteHelper v4

New features

• In addition to fill-in-the-blank questions, the software now supports single-choice questions
• Integrates an intelligent review feature that automatically recommends the questions you have least mastered, based on models and algorithms
• Add a preset exam feature to make generating exams more convenient
• Add some fun, energizing little features

Enhancements

• Optimize the quality of question generation by large models
• Refactor the question generation process to generate questions more effectively
• Optimize the UI interface
• Incorporate the SBERT large-scale model to improve matching accuracy
• Optimize the display of project creation progress

Bug Fixes

• Fixed an issue where the software failed to run on older system versions (21H2 and earlier) due to a lack of CET support
• Fixed a runtime crash caused by the missing vocab.txt file
• Fixed an issue where the software generated declarative statements, making it impossible to answer the questions
• Fixed an issue where configuration items failed to load

Build / Test updates

• The software has been completely refactored, so the project that now needs to be compiled is the WPF project
• The OpenAI package have been introduced to provide knowledge base generation function
• The relative project reference for AquaAvgEngine has been correctly re-established(#4)

Changelog

Full Changelog: v3...v4

The sky is vast, the setting sun seems far; The water’s clean, cold waves flow o’er the bar. Welcome to ReciteHelper!

ReciteHelper v3 "Plum"

ReciteHelper v3

New features

• Completely redesigned the project creation interface
• Provides a multi-file merging function, allowing easily combine multiple files
• Integrate with AquaAvgEngine, the question bank can be transformed into a visual novel game, increasing its entertainment value

Enhancements

• Optimize the progress indicator interface for creating traditional projects
• Introducing a replay algorithm to solve the problem of processing large files
• Integrate miscellaneous functions into the function menu
• Game chapters are created in parallel to speed up the process
• By introducing a runtime-time dynamic side-loading mechanism, the call price is reduced by 83%

Bug Fixes

• Fixed an issue where unremoved test code prevented file merging
• Fixed the issue of chapter re-clustering failure
• Fixed the issue where the software crashed due to being unable to read PDF files
• Fixed the software crash issue caused by the absence of necessary files for the jieba library

Build / Test updates

• The FreeSpire.Doc and FreeSpire.Presentation packages have been introduced to provide file merging support
• Added a project reference to AquaAvgEngine
• Refactor the compilation process

Changelog

Full Changelog: v2...v3

Welcome to ReciteHelper!

ReciteHelper v2f

ReciteHelper v2f

ReciteHelper v2f is a bug-fix version of ReciteHelper v2, addressing several issues that affected the software's performance and user experience.

Bug Fixes

• Fixed an issue where button position misalignment prevented users from clicking the export project button
• Fixed an issue where processing long PDFs caused errors due to AI hallucinations
• Fixed an issue where an error occurred during problem clustering due to some knowledge points being empty
• Fixed potential political risks in the sidebar

Welcome to ReciteHelper!

ReciteHelper v2 "Epiphyllum"

ReciteHelper v2

New features

• Added project import/export functionality
• Configuration items and accompanying configuration syntax parsing functionality
• The mock exam feature can now be customized
• Implemented knowledge point learning function
• Exam questions can now be reviewed and reports exported

Enhancements

• The block clustering algorithm has been optimized, reducing the time required to create projects
• Mock exams can be conducted by drawing questions according to the proportion of each chapter
• The user documentation has been updated to a new version
• Establish better workflows to solve user-raised issues

Bug Fixes

• Fixed the issue where no information was displayed when the software crashed and exited
• Fixed the issue where only one recently used project that no longer exists could be deleted at a time
• The software will no longer crash when exporting projects
• Fixed the issue of being unable to open a specified directory

Build / Test updates

• The Microsoft.CodeAnalysis.CSharp.Scripting (Roslyn API) ,Markdig.Wpf and Docnet.Core have been introduced to optimize the user experience
• Updated LlmTornado version references
• Introduced CI workflow with managed compilation and build upload

Changelog

Full Changelog: v1...v2

Welcome to ReciteHelper meow!

ReciteHelper v1 "Sunflower"

ReciteHelper v1

New features

• Automatically generate a question bank based on review materials
• Constant mock exam function
• Clustering and displaying the questions and chapters
• Answer record storage
• Answer sheet and answer status display

Enhancements

• The parallel partitioning algorithm effectively shortens the problem generation time
• The questions were further divided into different levels
• Using LCS, similarity coefficient, and cosine similarity has solved the answer recognition problem to some extent
• The clustering expansion algorithm has been optimized, resulting in lower complexity

Bug Fixes

• Fixed the issue of chapter progress showing as 0
• Introducing a weighting mechanism to prevent cosine similarity from performing poorly on short texts
• Fixed an issue with incorrect dependency properties for the "Agree Rules" checkbox
• Fixed an issue that caused the software to crash when opening older projects
• The software crash issue during word segmentation has been resolved

Build / Test updates

• Updated to .NET 10 framework
• Integration testing was conducted on the AnyCPU environment running Windows 11
• Added build references for Docnet.Core, FuzzyString, JiebaNet, and LlmTornado
• Updated LlmTornado version references

Welcome to ReciteHelper meow!