Update sentence-transformers requirement from ~=5.2.3 to ~=5.3.0 (#286)

Updates the requirements on
[sentence-transformers](https://github.com/huggingface/sentence-transformers)
to permit the latest version.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/huggingface/sentence-transformers/releases">sentence-transformers's
releases</a>.</em></p>
<blockquote>
<h2>v5.3.0 - Improved Contrastive Learning, New Losses, and Transformers
v5 Compatibility</h2>
<p>This minor version brings several improvements to contrastive
learning: <code>MultipleNegativesRankingLoss</code> now supports
alternative InfoNCE formulations (symmetric, GTE-style) and optional
hardness weighting for harder negatives. Two new losses are introduced,
<code>GlobalOrthogonalRegularizationLoss</code> for embedding space
regularization and <code>CachedSpladeLoss</code> for memory-efficient
SPLADE training. The release also adds a faster hashed batch sampler,
fixes <code>GroupByLabelBatchSampler</code> for triplet losses, and
ensures full compatibility with the latest Transformers v5 versions.</p>
<p>Install this version with</p>
<pre lang="bash"><code># Training + Inference
pip install sentence-transformers[train]==5.3.0
<h1>Inference only, use one of:</h1>
<p>pip install sentence-transformers==5.3.0
pip install sentence-transformers[onnx-gpu]==5.3.0
pip install sentence-transformers[onnx]==5.3.0
pip install sentence-transformers[openvino]==5.3.0
</code></pre></p>
<h2>Updated MultipleNegativesRankingLoss (a.k.a. InfoNCE)</h2>
<p><code>MultipleNegativesRankingLoss</code> received two major
upgrades: support for alternative InfoNCE formulations from the
literature, and optional hardness weighting to up-weight harder
negatives.</p>
<h3>Support other InfoNCE variants (<a
href="https://redirect.github.com/huggingface/sentence-transformers/issues/3607">#3607</a>)</h3>
<p><code>MultipleNegativesRankingLoss</code> now supports several
well-known contrastive loss variants from the literature through new
<code>directions</code> and <code>partition_mode</code> parameters.
Previously, this loss only supported the standard forward direction
(query → doc). You can now configure which similarity interactions are
included in the loss:</p>
<ul>
<li><code>&quot;query_to_doc&quot;</code> (default): For each query, its
matched document should score higher than all other documents.</li>
<li><code>&quot;doc_to_query&quot;</code>: The symmetric reverse — for
each document, its matched query should score higher than all other
queries.</li>
<li><code>&quot;query_to_query&quot;</code>: For each query, all other
queries should score lower than its matched document.</li>
<li><code>&quot;doc_to_doc&quot;</code>: For each document, all other
documents should score lower than its matched query.</li>
</ul>
<p>The <code>partition_mode</code> controls how scores are normalized:
<code>&quot;joint&quot;</code> computes a single softmax over all
directions, while <code>&quot;per_direction&quot;</code> computes a
separate softmax per direction and averages the losses.</p>
<p>These combine to reproduce several loss formulations from the
literature:</p>
<p><strong>Standard InfoNCE</strong> (default, unchanged behavior):</p>
<pre lang="python"><code>loss = MultipleNegativesRankingLoss(model)
# equivalent to directions=(&quot;query_to_doc&quot;,),
partition_mode=&quot;joint&quot;
</code></pre>
<p><strong>Symmetric InfoNCE</strong> (<a
href="https://arxiv.org/abs/2310.19923">Günther et al. 2024</a>) — adds
the reverse direction so both queries and documents are trained to find
their match:</p>
<pre lang="python"><code>loss = MultipleNegativesRankingLoss(
    model,
    directions=(&quot;query_to_doc&quot;, &quot;doc_to_query&quot;),
    partition_mode=&quot;per_direction&quot;,
)
</code></pre>
<p><strong>GTE improved contrastive loss</strong> (<a
href="https://arxiv.org/abs/2308.03281">Li et al. 2023</a>) — adds
same-type negatives (query &lt;-&gt; query, doc &lt;-&gt; doc) for a
stronger training signal, especially useful with pairs-only data:</p>
<pre lang="python"><code>loss = MultipleNegativesRankingLoss(
&lt;/tr&gt;&lt;/table&gt; 
</code></pre>
</blockquote>
<p>... (truncated)</p>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="https://github.com/huggingface/sentence-transformers/commit/ce48ecc77f8c7ed97c8fcc3cd99fde6071206680"><code>ce48ecc</code></a>
Merge branch 'main' into v5.3-release</li>
<li><a
href="https://github.com/huggingface/sentence-transformers/commit/cec08f89aabaa11a1675529fb7b11618c8255622"><code>cec08f8</code></a>
Fix citation for EmbeddingGemma paper (<a
href="https://redirect.github.com/huggingface/sentence-transformers/issues/3687">#3687</a>)</li>
<li><a
href="https://github.com/huggingface/sentence-transformers/commit/c29b3a69f7eca78d613fc2e465989ee880216c8d"><code>c29b3a6</code></a>
Release v5.3.0</li>
<li><a
href="https://github.com/huggingface/sentence-transformers/commit/55c13de47ea9475b1416a34af36716cc477bd95f"><code>55c13de</code></a>
Prep docs main page for v5.3.0 (<a
href="https://redirect.github.com/huggingface/sentence-transformers/issues/3686">#3686</a>)</li>
<li><a
href="https://github.com/huggingface/sentence-transformers/commit/72e75f7868700a768f6c57fb16e76b18df250f72"><code>72e75f7</code></a>
[<code>tests</code>] Add slow reproduction tests for most common models
(<a
href="https://redirect.github.com/huggingface/sentence-transformers/issues/3681">#3681</a>)</li>
<li><a
href="https://github.com/huggingface/sentence-transformers/commit/237e44151758a9b4eb86b1ecd73152884073549a"><code>237e441</code></a>
[<code>fix</code>] Fix model card generation with set_transform with new
column names (#...</li>
<li><a
href="https://github.com/huggingface/sentence-transformers/commit/7f180b4b7146af66e2b043cdaa89bf87babe28bc"><code>7f180b4</code></a>
[<code>feat</code>] Add hardness-weighted contrastive learning to losses
(<a
href="https://redirect.github.com/huggingface/sentence-transformers/issues/3667">#3667</a>)</li>
<li><a
href="https://github.com/huggingface/sentence-transformers/commit/58900864d51cde2f918885587709b2466938422c"><code>5890086</code></a>
Disallow query_to_query/doc_to_doc with
partition_mode=&quot;per_direction&quot; due to...</li>
<li><a
href="https://github.com/huggingface/sentence-transformers/commit/6518c36671294b2125506f0d382d04102f739c91"><code>6518c36</code></a>
CE trainer: Removed IterableDataset from train and eval dataset type
hints (#...</li>
<li><a
href="https://github.com/huggingface/sentence-transformers/commit/1e0e84cbfb5483ffe1a5ff7db459c67a61487f3f"><code>1e0e84c</code></a>
Add tips for adjusting batch size to improve processing speed (<a
href="https://redirect.github.com/huggingface/sentence-transformers/issues/3672">#3672</a>)</li>
<li>Additional commits viewable in <a
href="https://github.com/huggingface/sentence-transformers/compare/v5.2.3...v5.3.0">compare
view</a></li>
</ul>
</details>
<br />


Dependabot will resolve any conflicts with this PR as long as you don't
alter it yourself. You can also trigger a rebase manually by commenting
`@dependabot rebase`.

[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)

---

<details>
<summary>Dependabot commands and options</summary>
<br />

You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits
that have been made to it
- `@dependabot show <dependency name> ignore conditions` will show all
of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop
Dependabot creating any more for this major version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop
Dependabot creating any more for this minor version (unless you reopen
the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop
Dependabot creating any more for this dependency (unless you reopen the
PR or upgrade to it yourself)


</details>

Author: DefinetlyNotAI

requirements.txt

@@ -17,4 +17,4 @@ networkx~=3.6.1
 pandas~=3.0.1
 cryptography~=46.0.5
 scapy~=2.7.0
-sentence-transformers~=5.2.3
+sentence-transformers~=5.3.0