rootacess commited on
Commit
817653c
·
verified ·
1 Parent(s): e224a07

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -1
README.md CHANGED
@@ -16,6 +16,7 @@ datasets:
16
  - microsoft/NextCoderDataset
17
  - microsoft/NextCoderDataset-Conversational
18
  - bigcode/commitpackft
 
19
  ---
20
 
21
 
@@ -119,7 +120,7 @@ response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
119
 
120
  *Comparison of base QwenCoder-2.5 models of different sizes and their SELEKT-enhanced versions across three code editing benchmarks.*
121
 
122
- **Detailed evaluation results are reported in this [📑 paper](https://arxiv.org/abs/2503.03656).**
123
 
124
  ## Responsible AI Use
125
  The base models (from the QwenCoder-2.5 family) are suspectible to malicious prompts and may generate or execute harmful code. Our finetuning does not enhance or impede such behaviors. The users should use the models and their outputs responsibly and with caution. Model outputs should be subjected to additional analysis, including manual inspection, and sandboxing before execution.
 
16
  - microsoft/NextCoderDataset
17
  - microsoft/NextCoderDataset-Conversational
18
  - bigcode/commitpackft
19
+ - bigcode/starcoderdata
20
  ---
21
 
22
 
 
120
 
121
  *Comparison of base QwenCoder-2.5 models of different sizes and their SELEKT-enhanced versions across three code editing benchmarks.*
122
 
123
+ **Detailed evaluation results are reported in this [📑 paper](https://www.microsoft.com/en-us/research/publication/nextcoder-robust-adaptation-of-code-lms-to-diverse-code-edits/).**
124
 
125
  ## Responsible AI Use
126
  The base models (from the QwenCoder-2.5 family) are suspectible to malicious prompts and may generate or execute harmful code. Our finetuning does not enhance or impede such behaviors. The users should use the models and their outputs responsibly and with caution. Model outputs should be subjected to additional analysis, including manual inspection, and sandboxing before execution.