Downloads · 30 days
6
15% of all-time downloads
MaximeM/Llama-3.2-3B-recsys-challenge-2026-scorehead-top50
Llama-3.2-3B-recsys-challenge-2026-scorehead-top50 is a machine learning model from MaximeM. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as llama3.2.
Listwise scoring-head reranker for conversational music recommendation, from team FPMsUMONS's entry to the RecSys Challenge 2026. Trained on top-50 candidate pools. It is the shallow-pool comparison arm of the paper's…
Downloads · 30 days
6
15% of all-time downloads
All-time downloads
39
Public
Parameters
3.2B
6.4 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors6.4 GB · 100%
From the Hugging Face model README
Listwise scoring-head reranker for conversational music recommendation, from team FPMs_UMONS's entry to the RecSys Challenge 2026. Trained on top-50 candidate pools. It is the shallow-pool comparison arm of the paper's main table: training on a deeper pool measurably improves the deep ranking, which is one of the paper's findings.
This is a pipeline component, not a standalone model — and a comparison arm rather than the submitted system. It reranks a candidate pool produced by our retriever, with candidates rendered in one exact format and enriched with a descriptor cache. There is no standalone usage recipe here on purpose: the repository is what loads it correctly and runs it end to end.
MaximeM/Qwen3-8B-recsys-challenge-2026-scorehead-top200Evaluation numbers are reported in the paper and the repository; they are not duplicated here.
A marker token is appended after each candidate; the model reads that marker's hidden state in a single forward pass, and a linear head maps it to a score, trained with listwise cross-entropy. It ranks the whole pool at once rather than emitting a single pick — the distinction the paper is about.
config.json declares LlamaModel — the backbone without an LM head.
Loading with AutoModelForCausalLM does not give you this model.
The scoring head is not in the safetensors. It lives in score_head.pt,
which also records the marker token id; fetching only the safetensors gives a
silently unusable model.
Each candidate must be one line ending with the marker token
(<|reserved_special_token_5|>, id 128013), carrying the distilled
{sound | themes} descriptors the model was trained on:
12. Thunderstruck by AC/DC [rock, 80s] {sound: hard rock, metal | themes: rebellion, anger}<marker>
The descriptors are not optional, and the rendering is not approximate.
The repository handles all three, and the descriptor cache ships as a released artifact alongside the commands that rebuild it.
Built with Llama.
This model is a derivative of
meta-llama/Llama-3.2-3B-Instruct
and is therefore governed by the Llama 3.2 Community License, a copy of
which is included as LICENSE.txt, together with Meta's USE_POLICY.md.
Use of this model must comply with both.
Modifications: the transformer weights were adapted with LoRA (rank 16,
alpha 32) on the frozen base and merged; the original language-modelling head
was dropped and replaced by the linear scoring head in score_head.pt.