Add Mixed-Distill en-fa ColBERT-XLMR model (MiLQ, arXiv:2505.16631)

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,79 @@
+---
+language:
+- en
+- fa
+license: apache-2.0
+library_name: colbert
+base_model: FacebookAI/xlm-roberta-large
+pipeline_tag: feature-extraction
+tags:
+- colbert
+- late-interaction
+- dense-retrieval
+- cross-lingual
+- code-switching
+- information-retrieval
+---
+
+# Mixed-Distill enfa-fa (English–Persian)
+
+A cross-lingual **ColBERT** late-interaction retriever (XLM-RoBERTa-large backbone) for
+**English ⇄ Persian (Farsi)** web search. The model is **distilled** from a strong reranker and trained on
+**code-switched (mixed-language) queries**, following the *Mixed-Distill* recipe from the MiLQ paper.
+
+The repo name encodes the training direction: **`enfa-fa`** = code-switched **en+fa queries → fa documents**.
+It is designed to be robust when bilingual users issue **mixed English+Persian queries** against
+Persian (Farsi)-language documents.
+
+## What "Mixed-Distill" means
+
+- **Mixed** — queries are *code-switched* (English tokens randomly mixed into Persian queries, MUSE-based,
+  ~0.5 mixing ratio), so the model handles native, English, and mixed-language queries.
+- **Distill** — trained with knowledge distillation (KL-divergence) from teacher relevance scores
+  (mT5-XXL / monoT5 over mMARCO), 6-way passage scoring.
+
+## Intended use
+
+- **Queries:** English, Persian, or **code-switched English+Persian**.
+- **Documents:** Persian (Farsi)-language passages.
+- **Scoring:** ColBERT late interaction (MaxSim over per-token embeddings).
+
+## Specs
+
+| | |
+|---|---|
+| Base model | `xlm-roberta-large` |
+| Architecture | ColBERT (late interaction) |
+| Projection dim | 128 |
+| Similarity | cosine |
+| Query max length | 32 |
+| Doc max length | 180 |
+| Training | KD (KLD), n-way 6, teacher: mT5-XXL/monoT5 on mMARCO |
+
+## Usage
+
+Load with the [ColBERT](https://github.com/stanford-futuredata/ColBERT) library:
+
+```python
+from colbert.modeling.checkpoint import Checkpoint
+from colbert.infra import ColBERTConfig
+
+ckpt = Checkpoint("<your-username>/ColBERT-XLMR-Mixed-Distill-enfa-fa",
+                  colbert_config=ColBERTConfig())
+Q = ckpt.queryFromText(["mixed English+Persian query ..."])
+D = ckpt.docFromText(["Persian (Farsi) document passage ..."])
+```
+
+## Citation
+
+```bibtex
+@misc{kim2025milqbenchmarkingirmodels,
+      title={MiLQ: Benchmarking IR Models for Bilingual Web Search with Mixed Language Queries},
+      author={Jonghwi Kim and Deokhyung Kang and Seonjeong Hwang and Yunsu Kim and Jungseul Ok and Gary Lee},
+      year={2025},
+      eprint={2505.16631},
+      archivePrefix={arXiv},
+      primaryClass={cs.IR},
+      url={https://arxiv.org/abs/2505.16631},
+}
+```

added_tokens.json

@@ -0,0 +1,4 @@
+{
+  "[unused0]": 250002,
+  "[unused1]": 250003
+}

artifact.metadata

@@ -0,0 +1,92 @@
+{
+    "query_token_id": "[unused0]",
+    "doc_token_id": "[unused1]",
+    "query_token": "[Q]",
+    "doc_token": "[D]",
+    "ncells": null,
+    "centroid_score_threshold": null,
+    "ndocs": null,
+    "only_approx": false,
+    "index_path": null,
+    "nbits": 1,
+    "kmeans_niters": 4,
+    "resume": false,
+    "max_sampled_pid": -1,
+    "max_num_partitions": -1,
+    "use_lagacy_build_ivf": false,
+    "reuse_centroids_from": null,
+    "similarity": "cosine",
+    "bsize": 16,
+    "accumsteps": 1,
+    "lr": 5e-6,
+    "maxsteps": 200000,
+    "save_every": null,
+    "resume_optimizer": false,
+    "fix_broken_optimizer_state": false,
+    "warmup": null,
+    "warmup_bert": null,
+    "relu": false,
+    "nway": 6,
+    "n_query_alternative": 1,
+    "use_ib_negatives": false,
+    "kd_loss": "KLD",
+    "reranker": false,
+    "distillation_alpha": 1.0,
+    "ignore_scores": false,
+    "model_name": "xlm-roberta-large",
+    "force_resize_embeddings": true,
+    "shuffle_passages": true,
+    "sampling_max_beta": 1.0,
+    "over_one_epoch": true,
+    "multilang": false,
+    "nolangreg": false,
+    "listwise_loss": null,
+    "max_epoch": 5,
+    "multiresource": false,
+    "num_parallel_queries": 0,
+    "num_parallel_collections": 0,
+    "run_tag": "enfa-fa_run1",
+    "maxsim_force": false,
+    "switching_prob": 0.5,
+    "switching_lang": null,
+    "softmax_temperature": 1.0,
+    "kld_weight": 1.0,
+    "mse_weight": 0.0,
+    "align_loss_weight": 1.0,
+    "only_exact_match": false,
+    "on_the_fly_switching": false,
+    "doc_lang": null,
+    "entropy_regularization": false,
+    "entropy_beta": 0.1,
+    "multi_query": false,
+    "query_contrastive_loss": false,
+    "query_contrastive_loss_weight": 0.1,
+    "query_maxlen": 32,
+    "attend_to_mask_tokens": false,
+    "interaction": "colbert",
+    "dim": 128,
+    "doc_maxlen": 180,
+    "mask_punctuation": true,
+    "checkpoint": "xlm-roberta-large",
+    "triples": "\/home\/jhkim980112\/workspace\/dataset\/msmarco\/mt5xxl-monot5-mmarco-engeng.jsonl",
+    "collection": "\/home\/jhkim980112\/workspace\/dataset\/mMARCO\/google_translations\/collections\/fa_collection_sockeye2-tm1.tsv",
+    "queries": "\/home\/jhkim980112\/workspace\/dataset\/cs_marco\/queries\/MUSE\/en_fa_queries_random_0.5.tsv",
+    "index_name": null,
+    "parallel_queries": null,
+    "parallel_collections": null,
+    "maxsim_mapping": null,
+    "lexicon": null,
+    "debug": false,
+    "overwrite": false,
+    "root": "\/home\/jhkim980112\/workspace\/final_model_checkpoint\/",
+    "experiment": "CSD_xlmr",
+    "index_root": null,
+    "name": "enfa-fa_run1\/64bat.6way",
+    "rank": 0,
+    "nranks": 4,
+    "amp": true,
+    "ivf_num_processes": 20,
+    "ivf_use_tempdir": false,
+    "ivf_merging_ways": 2,
+    "gpus": 4
+}

config.json

@@ -0,0 +1,28 @@
+{
+  "_name_or_path": "xlm-roberta-large",
+  "architectures": [
+    "HF_ColBERT"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "xlm-roberta",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "output_past": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.41.1",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 250004
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c216924c7b986033fa1f5195579bf5888c2ee294a72c55dc116daa9847a5359d
+size 2240142872

sentencepiece.bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
+size 5069051

special_tokens_map.json

@@ -0,0 +1,15 @@
+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:410ff101f2c7d6eb4f670a9410e9e27063dc10cd0c82fb1925ee779788d3036d
+size 17083106

tokenizer_config.json

@@ -0,0 +1,70 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250001": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250002": {
+      "content": "[unused0]",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "250003": {
+      "content": "[unused1]",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "XLMRobertaTokenizer",
+  "unk_token": "<unk>"
+}