ppsingh commited on
Commit
c320e3c
·
verified ·
1 Parent(s): 3626564

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +131 -1
README.md CHANGED
@@ -8,4 +8,134 @@ app_port: 3000
8
  pinned: false
9
  ---
10
 
11
- Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
8
  pinned: false
9
  ---
10
 
11
+ ## About EUDR Agent
12
+ The EU Deforestation Regulation (EUDR) requires companies to ensure that specific commodities placed on the EU market are deforestation-free and legally produced.
13
+
14
+ This AI-powered tool helps stakeholders:
15
+
16
+ - Understand EUDR compliance requirements
17
+ - Analyze geographic deforestation data using WHISP API
18
+ - Assess supply chain risks
19
+ - Navigate complex regulatory landscapes
20
+ - Developed by GIZ to enhance accessibility and understanding of EUDR requirements through advanced AI and geographic data processing capabilities.
21
+
22
+ Key Features:
23
+ - Automatic analysis of uploaded GeoJSON files via WHISP API
24
+ - Country-specific EUDR compliance guidance
25
+ - Real-time question answering with source citations
26
+ - User-friendly interface for complex regulatory information
27
+
28
+ ----------------------------------------------------------------------------
29
+ ### 💬 How to Ask Effective Questions
30
+ ❌ Less Effective ✅ More Effective
31
+ "What is deforestation?" "What are the main deforestation hotspots in Ecuador?"
32
+ "Tell me about compliance" "What EUDR requirements apply to coffee imports from Guatemala?"
33
+ "Show me data" "What is the deforestation rate in the uploaded region?"
34
+
35
+ 🔍 Using Data Sources
36
+ Upload GeoJSON: Upload your geographic data files for automatic analysis via WHISP API
37
+ Talk to Reports: Select Ecuador or Guatemala for country-specific EUDR analysis
38
+
39
+ ### ⭐ Best Practices
40
+ - Be specific about regions, commodities, or time periods
41
+ - Ask one question at a time for clearer answers
42
+ - Use follow-up questions to explore topics deeper
43
+ - Provide context when possible
44
+
45
+ ---------------------------------------------------------------
46
+ ## Important Disclaimers
47
+ ⚠️ Scope & Limitations:
48
+
49
+ This tool is designed for EUDR compliance assistance and geographic data analysis
50
+ Responses should not be considered official legal or compliance advice
51
+ Always consult qualified professionals for official compliance decisions
52
+
53
+ ⚠️ Data & Privacy:
54
+ - Uploaded GeoJSON files are processed via external WHISP API for analysis
55
+ - We collect usage statistics to improve the tool
56
+ - Files are processed temporarily and not permanently stored
57
+
58
+ ⚠️ AI Limitations:
59
+
60
+ - Responses are AI-generated and may contain inaccuracies
61
+ - The tool is a prototype under continuous development
62
+ - Always verify important information with authoritative sources
63
+
64
+ Data Collection: We collect questions, answers, feedback, and anonymized usage statistics to improve tool performance based on legitimate interest in service enhancement.By using this chatbot, you agree to these terms and acknowledge that you are solely responsible for any reliance on or actions taken based on its responses.
65
+
66
+ Technical Information: User can read more about the technical information about the tool in section below.
67
+
68
+ This is just a prototype and being tested and worked upon, so its not perfect and may sometimes give irrelevant answers. If you are not satisfied with the answer, please ask a more specific question or report your feedback to help us improve the system.
69
+
70
+
71
+ -----------------------------------------------------------------------------------
72
+
73
+
74
+ # Technical Documentation of the system in accordance with EU AI Act
75
+
76
+ **System Name:** EUDR Chatbot
77
+
78
+ **Provider / Supplier:** GIZ Data Service Center
79
+
80
+ **As of:** September 2025
81
+
82
+ ## 1. General Description of the System
83
+
84
+ EUDR Bot is an AI-powered conversational assistant designed to help you understand compliance with and analyze the EU Deforestation Regulation. This tool leverages advanced language models to help you get clear and structured answers about EUDR requirements, compliance procedures, and regulatory guidance.
85
+
86
+ It combines a generative language assistant with a knowledge base implemented via Retrieval-Augmented Generation (RAG). In addition to the RAG, the tool also provide quick analysis on the geojson of plot by leveraging the [Whisp](https://openforis.org/solutions/whisp/). The scope and functionality of the tool is focused on EU Deforestation Regulation compliance and related documentation in context of 'Ecuador' and 'Guatemala' only.
87
+
88
+ ## 2. Models Used
89
+
90
+ ### Generative LLM
91
+ - **Model Name:** [meta-llama/Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct)
92
+ - **Model Source API:** [Nebius AI](https://studio.nebius.com/)
93
+
94
+ ### Retriever/Embedding
95
+ - **Model Name:** [BAAI/bge-m3](https://huggingface.co/BAAI/bge-m3)
96
+ - **Model Source:** Local Instance
97
+
98
+ ### Re-ranker
99
+ - **Model Name:** [BAAI/bge-reranker-v2-m3](https://huggingface.co/BAAI/bge-reranker-v2-m3)
100
+ - **Model Source:** Local Instance
101
+
102
+ ### External API's
103
+ - **Whisp Info**: https://openforis.org/solutions/whisp/
104
+ - **API Endpoint** https://whisp.openforis.org/documentation/api-guide
105
+
106
+ ## 3. Model Training Data
107
+
108
+ All the models mentioned above are being consumed without any fine-tuning or training being performed by the developer team of EUDR Bot. And hence there is no training data which had been used by the development team of EUDR Bot.
109
+
110
+ ## 4. Knowledge Base (Retrieval Component)
111
+
112
+ - **Data Sources:** Public EUDR documentation, regulatory guidance, and compliance materials
113
+ - **Embedding Model:** [BAAI/bge-m3](https://huggingface.co/BAAI/bge-m3)
114
+ - **Embedding Dimension:** 1024
115
+ - **Vector Database:** Qdrant (via API)
116
+ - **Framework:** Langchain (custom RAG pipeline)
117
+ - **Top-k:** 10 relevant text segments per query
118
+
119
+ ## 5. System Limitations and Non-Purposes
120
+
121
+ - The system does not make autonomous decisions.
122
+ - No processing of personal data except for the usage statistics as mentioned in Disclaimer.
123
+ - Results are intended for orientation only – not for legal or regulatory compliance advice.
124
+ - Users should consult official EU documentation and legal experts for definitive compliance guidance.
125
+
126
+ ## 6. Transparency Towards Users
127
+
128
+ - The user interface clearly indicates the use of a generative AI model.
129
+ - An explanation of the RAG method is included.
130
+ - We collect usage statistics as detailed in Disclaimer tab of the app along with the explicit display in the user interface of the tool.
131
+ - Feedback mechanism available (via https://huggingface.co/spaces/GIZ/eudr_chatbo_chatui/discussions/new).
132
+
133
+ ## 7. Monitoring, Feedback, and Incident Reporting
134
+
135
+ - User can provide feedback via UI by giving (Thumbs-up or down to AI-Generated answer). Alternatively for more detailed feedback please use https://huggingface.co/spaces/GIZ/eudr_chatbo_chatui/discussions/new to report any issue.
136
+ - Technical development is carried out by the GIZ Data Service Center.
137
+ - No automated bias detection – but low risk due to content restrictions.
138
+
139
+ ## 8. Contact
140
+
141
+ For any questions, please contact via https://huggingface.co/spaces/GIZ/eudr_chatbo_chatui/discussions/new or send us email to [email protected]